![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F443651%2F178ad5d3d446a118de0825df1fd5c558%2FScreenshot%20from%202020-08-17%2011-06-21.png?generation=1597656073961549&alt=media\")
I got contacted by a source that prefers to stay anonymous with some very valuable information that I would like to make public.
\nWhy are people private sharing?
\nOne could ask why people share their notebooks or submission files and decrease their own odds of winning. The answer is simple: money. There are organizations out there that sell Kaggle medals for quite a large amount of money. Here's an example of such a store. Here, a guaranteed 1% finish (silver medal) is being sold for 1000 yuan (apparently that has to be paid 4 times which results in 600 USD).
Here's another screenshot where they boast with how 2 of their students got to master tier and how they secured 38 (!!) silver medals within one competition. What a success rate!
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F443651%2F23e6904397bd40a985afd05595fe6211%2FNPk9inL1.jpg%20medium.jpeg?generation=1597656104553287&alt=media\")
Here's a rough outline of how the process goes:
\nWho is behind this?
\nAdvertisements that include the names of two kaggle masters can be found:
![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F443651%2F2b4d67b09ebaff8e8b446afd6ac2f48d%2FqAa3caEF.jpg%20medium.jpeg?generation=1597656299456036&alt=media\")
![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F443651%2F4a944e66287b82a4c8b4d8476c2ddd23%2FUpa2ksgG.jpeg?generation=1597656322742858&alt=media\")
These are screenshots from different competitions. What a coincidence, but these two accounts are currently ranked 18th on the public LB as well. Not too far away from all the fake accounts on the top. The chinese name of the first master is 王苏宏, which is also used to advertise 20 different accounts in the deepfake competition:
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F443651%2F7e8c653bc3fba6be0ea14c996b9e8428%2FPiOltSmI.jpg%20medium.jpeg?generation=1597656467503434&alt=media\")
What else can we do?
\nWe need to identify all accounts related to this suspicious cluster. One way to do this would be to create a graph analysis based on the Kaggle meta dataset.
I would also like to clarify that I might have blood on my hands. During this competition, I teamed up with someone that seemed very trust-worthy at first glance based on previous competition results & current LB standing. After teaming up, this person went completely silent and we never heard from him/her again. We also suspect the profile picture used to be fake (as it can be found on other instances with Google Images). I'll clarify this further after the competition to avoid any kind of griefing.
", "messages": [ { "id": 973472, "postDate": "2020-08-17T10:16:21.853Z", "content": "I got contacted by a source that prefers to stay anonymous with some very valuable information that I would like to make public.
\nWhy are people private sharing?
\nOne could ask why people share their notebooks or submission files and decrease their own odds of winning. The answer is simple: money. There are organizations out there that sell Kaggle medals for quite a large amount of money. Here's an example of such a store. Here, a guaranteed 1% finish (silver medal) is being sold for 1000 yuan (apparently that has to be paid 4 times which results in 600 USD).
Here's another screenshot where they boast with how 2 of their students got to master tier and how they secured 38 (!!) silver medals within one competition. What a success rate!
\n
Here's a rough outline of how the process goes:
\nWho is behind this?
\nAdvertisements that include the names of two kaggle masters can be found:
These are screenshots from different competitions. What a coincidence, but these two accounts are currently ranked 18th on the public LB as well. Not too far away from all the fake accounts on the top. The chinese name of the first master is 王苏宏, which is also used to advertise 20 different accounts in the deepfake competition:
\n
What else can we do?
\nWe need to identify all accounts related to this suspicious cluster. One way to do this would be to create a graph analysis based on the Kaggle meta dataset.
I would also like to clarify that I might have blood on my hands. During this competition, I teamed up with someone that seemed very trust-worthy at first glance based on previous competition results & current LB standing. After teaming up, this person went completely silent and we never heard from him/her again. We also suspect the profile picture used to be fake (as it can be found on other instances with Google Images). I'll clarify this further after the competition to avoid any kind of griefing.
", "rawMarkdown": "I got contacted by a source that prefers to stay anonymous with some very valuable information that I would like to make public.\n\n**Why are people private sharing?**\nOne could ask why people share their notebooks or submission files and decrease their own odds of winning. The answer is simple: money. There are organizations out there that sell Kaggle medals for quite a large amount of money. Here's an [example of such a store](https://shop319532120.world.taobao.com/?spm=a230r.7195193.1997079397.2.52a85a59xgPAS6). Here, a guaranteed 1% finish (silver medal) is being sold for 1000 yuan (apparently that has to be paid 4 times which results in 600 USD).\n\n\n\nHere's another screenshot where they boast with how 2 of their students got to master tier and how they secured 38 (!!) silver medals within one competition. What a success rate!\n\n\nHere's a rough outline of how the process goes:\n1. students find them on websites or shops(taobao淘宝)\n2. they talk to the front desk about what's going on, asking for prices and services included\n3. they have different prices for different medals, for instance, if you want a silver, you need to pay more\n4. their lessons do not provide anything meaningful to anyone, using it as a mask for their true business\n5. you will also be assigned to a teacher where he answers your questions really really poorly.\n6. in kernel only competitions, they will give you models. In csv competitions, they will give you subs(for instance in SIIM)\n7. after you finish this competition, you learned nothing but how to cheat\n\n**Who is behind this?**\nAdvertisements that include the names of two kaggle masters can be found:\n\n\n\n\n\nThese are screenshots from different competitions. What a coincidence, but these two accounts are currently ranked 18th on the public LB as well. Not too far away from all the fake accounts on the top. The chinese name of the first master is 王苏宏, which is also used to advertise 20 different accounts in the deepfake competition:\n\n\n**What else can we do?**\nWe need to identify all accounts related to this suspicious cluster. One way to do this would be to create a graph analysis based on the [Kaggle meta dataset](https://www.kaggle.com/kaggle/meta-kaggle).\n\n**I would also like to clarify that I might have blood on my hands. During this competition, I teamed up with someone that seemed very trust-worthy at first glance based on previous competition results & current LB standing. After teaming up, this person went completely silent and we never heard from him/her again. We also suspect the profile picture used to be fake (as it can be found on other instances with Google Images). I'll clarify this further after the competition to avoid any kind of griefing.**", "votes": 341 }, { "id": 973631, "postDate": "2020-08-17T12:14:17.677Z", "content": "Almost all the golden area is overtaken by novice cheaters accounts.
\nIf they're still there on private LB, I will likely never compete again on kaggle.
", "rawMarkdown": "Almost all the golden area is overtaken by novice cheaters accounts. \n\nIf they're still there on private LB, I will likely never compete again on kaggle. ", "votes": 44, "replies": [ { "id": 973662, "postDate": "2020-08-17T12:46:41.907Z", "content": "the [angtk.ai]poxiao team was not removed. they simply dropped 11 spots to finish 15 on private LB and renamed to their team to oxiao.
", "rawMarkdown": "the [angtk.ai]poxiao team was not removed. they simply dropped 11 spots to finish 15 on private LB and renamed to their team to oxiao. ", "votes": 3 }, { "id": 973669, "postDate": "2020-08-17T12:53:26.957Z", "content": "Meant for the ongoing Melanoma competition.
", "rawMarkdown": "Meant for the ongoing Melanoma competition.", "votes": 1 }, { "id": 973741, "postDate": "2020-08-17T13:44:20.153Z", "content": "That is not right. You cannot commit suicide just because you don’t like people around you. Stay with it and we’ll have a solution together, or at least there are people who support you.
", "rawMarkdown": "That is not right. You cannot commit suicide just because you don’t like people around you. Stay with it and we’ll have a solution together, or at least there are people who support you. ", "votes": 11 }, { "id": 973789, "postDate": "2020-08-17T14:14:32.640Z", "content": "Your comparison does not seem very appropriate to me.
\nThey are apparently serial cheaters who have been doing it for a long time. Faced with such a body of evidences gathered and shared here, if kaggle doesn't sanction them, I don't see what would keep them from doing it over and over again, and even inciting others to do the same given there would be no sanction. And the competitions will lose all their interest.
", "rawMarkdown": "Your comparison does not seem very appropriate to me. \n\nThey are apparently serial cheaters who have been doing it for a long time. Faced with such a body of evidences gathered and shared here, if kaggle doesn't sanction them, I don't see what would keep them from doing it over and over again, and even inciting others to do the same given there would be no sanction. And the competitions will lose all their interest.", "votes": 5 }, { "id": 973815, "postDate": "2020-08-17T14:26:58.297Z", "content": "Just be here, don't go away. You are a passionate data scientist so don't flee away from anything
", "rawMarkdown": "Just be here, don't go away. You are a passionate data scientist so don't flee away from anything", "votes": 3 } ] }, { "id": 973698, "postDate": "2020-08-17T13:14:15.740Z", "content": "I hope Kaggle will ban all of those who buy or sell medals.
", "rawMarkdown": "I hope Kaggle will ban all of those who buy or sell medals.", "votes": 30 }, { "id": 974218, "postDate": "2020-08-17T19:54:20.337Z", "content": "You have a future in Intelligence industry ;)
", "rawMarkdown": "You have a future in Intelligence industry ;)", "votes": 26, "replies": [ { "id": 974219, "postDate": "2020-08-17T19:56:13.757Z", "content": "No word from the Belgian government yet unfortunately… They might be too busy forming a government.
", "rawMarkdown": "No word from the Belgian government yet unfortunately... They might be too busy forming a government.", "votes": 30 }, { "id": 974249, "postDate": "2020-08-17T20:34:10.970Z", "content": "Are they still choosing the government? Is that another 2010-11 crisis? :D
\nhttps://en.wikipedia.org/wiki/2010%E2%80%9311_Belgian_government_formation
We broke our own world record :)
\n\n", "rawMarkdown": "We broke our own world record :)\n\nhttps://www.brusselstimes.com/belgium/124777/belgium-breaks-own-record-for-longest-period-without-government/\n\nhttps://www.guinnessworldrecords.com/world-records/96893-longest-time-without-a-government-in-peacetime", "votes": 2 }, { "id": 974267, "postDate": "2020-08-17T20:56:19.333Z", "content": "That's both funny and sad. Vive la Belgique. :)
", "rawMarkdown": "That's both funny and sad. Vive la Belgique. :)", "votes": 2 }, { "id": 974283, "postDate": "2020-08-17T21:11:06.993Z", "content": "At least you have fries ;)
", "rawMarkdown": "At least you have fries ;)", "votes": 2 }, { "id": 974286, "postDate": "2020-08-17T21:13:03.043Z", "content": "And gaufre. 😍
", "rawMarkdown": "And gaufre. 😍", "votes": 2 }, { "id": 974299, "postDate": "2020-08-17T21:20:07.767Z", "content": "Hopefully Americans don't call them \"french gaufre\" (like french fries ^^)
\nBut I have not much hope for that , given for Trump : \"Belgium is a beautiful city\" :)
", "rawMarkdown": "Hopefully Americans don't call them \"french gaufre\" (like french fries ^^)\n\nBut I have not much hope for that , given for Trump : [\"Belgium is a beautiful city\"](https://www.youtube.com/watch?v=hIU8cAp-2VM) :)", "votes": 1 }, { "id": 974307, "postDate": "2020-08-17T21:28:12.283Z", "content": "Well, given its size… 😏
", "rawMarkdown": "Well, given its size... 😏 ", "votes": 1 }, { "id": 975393, "postDate": "2020-08-18T09:18:53.087Z", "content": "Petit mais parfait! Don't be jealous of our national treasures guys: beers, waffles, chocolate, fries, …
\nYou french people still have your fermented grape juice and moldy milk!
", "rawMarkdown": "Petit mais parfait! Don't be jealous of our national treasures guys: beers, waffles, chocolate, fries, ...\n\nYou french people still have your fermented grape juice and moldy milk!", "votes": 4 }, { "id": 975411, "postDate": "2020-08-18T09:32:24.830Z", "content": "I just searched \"10 things Belgium is famous for\", the first thing was the highest number of Chocolate sales of course, but the second one is this: 😄😄
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2070357%2F1281b2ad81c615a9b235d71bcff86752%2FScreenshot_2020-08-18%2010%20Things%20Belgium%20Is%20Famous%20for.png?generation=1597743099625456&alt=media\")
I didn't know it was that big of a achievement lol!
", "rawMarkdown": "I just searched \"10 things Belgium is famous for\", the first thing was the highest number of Chocolate sales of course, but the second one is this: 😄😄\n\n\n\nI didn't know it was that big of a achievement lol!", "votes": 2 }, { "id": 975532, "postDate": "2020-08-18T10:42:34.523Z", "content": "It's one of our greatest national prides. A country as small as ours doesn't get to seize a world record every day… Do you have the article? I'm intrigued by point 3 as well (party freaks, which are references to Tomorrowland I guess)
\nEDIT: Also: \"Everything ran smoothly during this era\" 🙈🙈🙈
", "rawMarkdown": "It's one of our greatest national prides. A country as small as ours doesn't get to seize a world record every day... Do you have the article? I'm intrigued by point 3 as well (party freaks, which are references to Tomorrowland I guess)\n\nEDIT: Also: \"Everything ran smoothly during this era\" 🙈🙈🙈", "votes": 2, "replies": [ { "id": 975573, "postDate": "2020-08-18T11:09:34.677Z", "content": "@group16 Everything ran smoothly during that era . Ofcourse ! Point 1 is chocolate and point 3 is party freak ! Nothing can go wrong there :) People are happy ..
", "rawMarkdown": "@group16 Everything ran smoothly during that era . Ofcourse ! Point 1 is chocolate and point 3 is party freak ! Nothing can go wrong there :) People are happy ..", "votes": 1 } ] }, { "id": 975576, "postDate": "2020-08-18T11:12:11.413Z", "content": "Ohh God, Tomorrow land is from Belgium? I didn't knew that either, what am I doing in my life. :/
\nCool, here is your Pride. 😄
\nBtw, in our state, our govt also moved to a hotel for some picnic stuff for a month and we were without any government too, although that's a different thing that we don't want them either! 🤕🤕
\nP.S.: I read the article again, Belgium really is amazing!
", "rawMarkdown": "Ohh God, Tomorrow land is from Belgium? I didn't knew that either, what am I doing in my life. :/\n\nCool, here is your [Pride](https://lets-travel-more.com/10-things-belgium-is-famous-for/). 😄\n\nBtw, in our state, our govt also moved to a hotel for some picnic stuff for a month and we were without any government too, although that's a different thing that we don't want them either! 🤕🤕\n\n*****\nP.S.: I read the article again, Belgium really is amazing!", "votes": 1 }, { "id": 975646, "postDate": "2020-08-18T11:57:24.957Z", "content": "That was a fun article, many thanks. I've drank some beers in 'T Brugs Beertje that can be seen in one of the pictures (which is located in my home-city).
", "rawMarkdown": "That was a fun article, many thanks. I've drank some beers in 'T Brugs Beertje that can be seen in one of the pictures (which is located in my home-city).", "votes": 1 } ] }, { "id": 973849, "postDate": "2020-08-17T15:00:56.650Z", "content": "Thank you, Gilles and your anonymous source, for exposing those fraudulent schemes. These methods constitute such a blatant disrespect of the community; it is very disappointing to see. I hope that the Kaggle team will take swift and decisive actions against them. The hard work and effort of all honest competitors should not be undermined like this.
\nTo Kaggle's credit, we know that accounts of cheaters are regularly removed, and that a lot of work is being done behind the scenes to keep the community clean. In the past, even cheating \"Grandmasters\" have been banned. However, those higher-profile cases normally take quite a lot of time, during which the bad actors can get away with more cheating. While I can understand that Kaggle wants to err on the side of caution, time is really of the essence here. A lot of damage is being done if groups of people can get away with fraud for even 1 competition, since that invites copycat behaviour.
\nAs Kaggle continues to grow in popularity, anti-fraud actions need to be taken faster. Otherwise our community might gradually lose its hard-earned reputation.
\nI would also like to say, Gilles, that you and your honest teammates did nothing wrong during this competition. You did your due diligence before teaming up, and in a better situation this shady account would likely have never had the results to make it seem trustworthy. I hope that Kaggle will find a way to punish the cheaters without it affecting the honest competitors.
", "rawMarkdown": "Thank you, Gilles and your anonymous source, for exposing those fraudulent schemes. These methods constitute such a blatant disrespect of the community; it is very disappointing to see. I hope that the Kaggle team will take swift and decisive actions against them. The hard work and effort of all honest competitors should not be undermined like this.\n\nTo Kaggle's credit, we know that accounts of cheaters are regularly removed, and that a lot of work is being done behind the scenes to keep the community clean. In the past, even cheating \"Grandmasters\" have been banned. However, those higher-profile cases normally take quite a lot of time, during which the bad actors can get away with more cheating. While I can understand that Kaggle wants to err on the side of caution, time is really of the essence here. A lot of damage is being done if groups of people can get away with fraud for even 1 competition, since that invites copycat behaviour.\n\n**As Kaggle continues to grow in popularity, anti-fraud actions need to be taken faster. Otherwise our community might gradually lose its hard-earned reputation.**\n\nI would also like to say, Gilles, that you and your honest teammates did nothing wrong during this competition. You did your due diligence before teaming up, and in a better situation this shady account would likely have never had the results to make it seem trustworthy. I hope that Kaggle will find a way to punish the cheaters without it affecting the honest competitors.", "votes": 24 }, { "id": 974013, "postDate": "2020-08-17T16:59:59.570Z", "content": "@divyansh22 if I see a cheater from your country, I would never say your country is full of cheaters.
\nPlease pay attention, we're talking about cheating not racism.
", "rawMarkdown": "@divyansh22 if I see a cheater from your country, I would never say your country is full of cheaters.\n\nPlease pay attention, we're talking about cheating not racism.", "votes": 22, "replies": [ { "id": 974724, "postDate": "2020-08-18T03:05:19.887Z", "rawMarkdown": "", "votes": 1, "isDeleted": true }, { "id": 974785, "postDate": "2020-08-18T03:52:05.507Z", "content": "I'm not surprised that some hateful words come from our neighbor…
", "rawMarkdown": "I'm not surprised that some hateful words come from our neighbor..." } ] }, { "id": 973497, "postDate": "2020-08-17T10:30:50.353Z", "content": "Fortunately I did not delete my WeChat app after Kaggle Days in Beijing. Let's go for shopping!
\nBtw I also have 11 silver medals and a few of them I don't need anymore if anyone interested :D
", "rawMarkdown": "Fortunately I did not delete my WeChat app after Kaggle Days in Beijing. Let's go for shopping!\n\nBtw I also have 11 silver medals and a few of them I don't need anymore if anyone interested :D", "votes": 18, "replies": [ { "id": 973501, "postDate": "2020-08-17T10:31:42.843Z", "content": "What's your offer? :D
", "rawMarkdown": "What's your offer? :D", "votes": 2 }, { "id": 973503, "postDate": "2020-08-17T10:34:51.863Z", "content": "You wanna put them on sale? 😂😂
", "rawMarkdown": "You wanna put them on sale? 😂😂" }, { "id": 973528, "postDate": "2020-08-17T10:50:25.013Z", "content": "", "rawMarkdown": "https://2.taobao.com/", "votes": 2 }, { "id": 974185, "postDate": "2020-08-17T19:24:52.690Z", "rawMarkdown": "", "isDeleted": true }, { "id": 974753, "postDate": "2020-08-18T03:29:04.543Z", "content": "Any \"Kaggler discount coupon\" for that? 😅
", "rawMarkdown": "Any \"Kaggler discount coupon\" for that? 😅", "replies": [ { "id": 974791, "postDate": "2020-08-18T03:54:57.090Z", "content": "@stevenyin I do have one discount code, but please don't share it, I had to pay for it. Its code TEAMYIKOUOUBAO and you can use it at any of angtk.com's various Kaggle markets.
", "rawMarkdown": "@stevenyin I do have one discount code, but please don't share it, I had to pay for it. Its code TEAMYIKOUOUBAO and you can use it at any of angtk.com's various Kaggle markets.", "votes": 2 } ] }, { "id": 974818, "postDate": "2020-08-18T04:11:24.720Z", "content": "Just tried it and only found already expired. I must be too slow to use it..☹️
", "rawMarkdown": "Just tried it and only found already expired. I must be too slow to use it..☹️" } ] }, { "id": 973790, "postDate": "2020-08-17T14:14:40.287Z", "content": "For those who’ve worked in the fraud detection space, it’s so interesting because transaction time (or submission time in our case) is such an interesting feature. Intuitively, you’d think they’d stagger their submissions to make the fraud seem more organic, but that goes against another driving factor, which is the desire to make as much money off the scheme as quickly as possible before the lid gets blown off the operation.
\n🍿🍿🍿🍿🍿
", "rawMarkdown": "For those who’ve worked in the fraud detection space, it’s so interesting because transaction time (or submission time in our case) is such an interesting feature. Intuitively, you’d think they’d stagger their submissions to make the fraud seem more organic, but that goes against another driving factor, which is the desire to make as much money off the scheme as quickly as possible before the lid gets blown off the operation.\n\n🍿🍿🍿🍿🍿", "votes": 15 }, { "id": 973736, "postDate": "2020-08-17T13:38:12.217Z", "content": "Nice work, @group16 . Kaggle needs you, the world needs you. :D
", "rawMarkdown": "Nice work, @group16 . Kaggle needs you, the world needs you. :D", "votes": 15 }, { "id": 974159, "postDate": "2020-08-17T18:57:57.700Z", "content": "On a lighter note, i think it's time we will soon see a kaggle competition about detecting fraud accounts.
", "rawMarkdown": "On a lighter note, i think it's time we will soon see a kaggle competition about detecting fraud accounts.", "votes": 16 }, { "id": 979291, "postDate": "2020-08-20T18:38:56.657Z", "content": "Update here:
\nhttps://www.kaggle.com/general/176169
", "rawMarkdown": "Update here:\n\nhttps://www.kaggle.com/general/176169", "votes": 14 }, { "id": 973624, "postDate": "2020-08-17T12:07:02.310Z", "content": "One thing I don't understand - WHY on Earth people pay 600-1000 $ for just a silver medal??
\nCorrect me if I'm wrong, but that's kinda a lot of money (well, maybe not if you're from US/EU) - just for example, that's brand new iPhone or a pretty solid PC or a week of vacation.
\nI remember someone wrote that it helps a lot when looking for jobs, but I seriously doubt it (especially in case of silver/bronze medals; it's understandable that gold may help in some way). Kaggle medals may be good in your resume but only if you already have some valuable experience. And, which matters most, if you're able to explain what you did to get those medals :D
TBH it also resembles buyers of loot and achievements in MMORPG/MOBA - just buying pixels to tickle one's sense of own importance.
", "rawMarkdown": "One thing I don't understand - WHY on Earth people pay 600-1000 $ for just a silver medal??\nCorrect me if I'm wrong, but that's kinda a lot of money (well, maybe not if you're from US/EU) - just for example, that's brand new iPhone or a pretty solid PC or a week of vacation.\nI remember someone wrote that it helps a lot when looking for jobs, but I seriously doubt it (especially in case of silver/bronze medals; it's understandable that gold may help in some way). Kaggle medals may be good in your resume but only if you already have some valuable experience. And, which matters most, if you're able to explain what you did to get those medals :D\n\nTBH it also resembles buyers of loot and achievements in MMORPG/MOBA - just buying pixels to tickle one's sense of own importance.", "votes": 13, "replies": [ { "id": 973756, "postDate": "2020-08-17T13:50:28.507Z", "content": "From the standpoint of a student, Kaggle medals are attractive. Student don't have any experience on projects and they don't know what's helpful to their job hunting. When they see a fancy ads that guarantees a job offer during their nervous job hunting period, it's likely that they take the shortcuts.
\nTheir kaggle path: get a medal -> learn -> (practice)
\nOur kaggle path: learn -> practice -> (get a medal)
Oh Kaggle medals are attractive for sure! I like Kaggle medals myself :D
\nTho I disagree with you - students have projects (course projects, scientific researches) and they know what's helpful (career fairs, networking, university advertisements for \"companies-partners\"). When I was in my grad year 2 years ago I already had 2 internships finished and some experience as translator and tech-writer. Is there really so much different situation in China?
\nI agree about shortcuts, but when one takes a shortcut, he should use his head and think how he will be using it. You buy stuff - you don't get to learn and practice - when you're asked about it - you're dead :)
Same for everywhere in the world. But you are talking about smart student, right? Apparently they are not. I can see they were not using their head and did not think from their $1000 charity.
", "rawMarkdown": "Same for everywhere in the world. But you are talking about smart student, right? Apparently they are not. I can see they were not using their head and did not think from their $1000 charity.", "votes": 1 }, { "id": 974771, "postDate": "2020-08-18T03:41:31.857Z", "content": "There are many things in the world you can't fully understand if you don't put your feet in their shoes. Just like I can't understand why women are addicted to diamonds. It's obviously a fraud and IQ-tax collector, from my point of perspective though, but you have to pay it if you want to marry her. Women don't believe diamonds can guarantee their husbands' love are eternal, but they still can't help to buy diamonds….That's my metaphor of kaggle medals to those cheaters.
", "rawMarkdown": "There are many things in the world you can't fully understand if you don't put your feet in their shoes. Just like I can't understand why women are addicted to diamonds. It's obviously a fraud and IQ-tax collector, from my point of perspective though, but you have to pay it if you want to marry her. Women don't believe diamonds can guarantee their husbands' love are eternal, but they still can't help to buy diamonds....That's my metaphor of kaggle medals to those cheaters. " } ] }, { "id": 973622, "postDate": "2020-08-17T12:04:58.987Z", "content": "Check these topics below.
\nhttps://www.kaggle.com/c/ashrae-energy-prediction/discussion/122203
\nhttps://www.kaggle.com/c/ieee-fraud-detection/discussion/110778
I wrote this post 8 months ago in the first topic and people instantly downvoted me and accused me of being racist.
\n\n\nHow do you explain other 100 novices from the same country in top 200 with very few submissions? Are their job also highly related to this competition?
\n
I still see the same pattern here. Some people get extremely defensive while trying to justify them.
", "rawMarkdown": "Check these topics below.\nhttps://www.kaggle.com/c/ashrae-energy-prediction/discussion/122203\nhttps://www.kaggle.com/c/ieee-fraud-detection/discussion/110778\n\nI wrote this post 8 months ago in the first topic and people instantly downvoted me and accused me of being racist.\n\n> How do you explain other 100 novices from the same country in top 200 with very few submissions? Are their job also highly related to this competition?\n\nI still see the same pattern here. Some people get extremely defensive while trying to justify them.\n\n\n", "votes": 13, "replies": [ { "id": 973623, "postDate": "2020-08-17T12:06:30.777Z", "content": "Same for the Liverpool competition where @authman pointed out some very suspicious behaviour too.
", "rawMarkdown": "Same for the Liverpool competition where @authman pointed out some very suspicious behaviour too.", "votes": 2 }, { "id": 973628, "postDate": "2020-08-17T12:11:54.740Z", "content": "Yes, I found that topic as well.
\nhttps://www.kaggle.com/c/liverpool-ion-switching/discussion/152417
\nGuess what? Someone was offended as usual.
", "rawMarkdown": "Yes, I found that topic as well.\n\nhttps://www.kaggle.com/c/liverpool-ion-switching/discussion/152417\n\nGuess what? Someone was offended as usual.", "votes": 4 }, { "id": 973633, "postDate": "2020-08-17T12:16:31.720Z", "content": "I was expecting (and prepared for) those kind of comments, but I've been lucky to not yet have received such bullshit :)
", "rawMarkdown": "I was expecting (and prepared for) those kind of comments, but I've been lucky to not yet have received such bullshit :)", "votes": 5 }, { "id": 973646, "postDate": "2020-08-17T12:27:42.587Z", "content": "\"数竞之家\" in ion switching submission is actually angtk.ai https://zhuanlan.zhihu.com/p/150559937
", "rawMarkdown": "\"数竞之家\" in ion switching submission is actually angtk.ai https://zhuanlan.zhihu.com/p/150559937", "votes": 2 }, { "id": 973661, "postDate": "2020-08-17T12:44:25.583Z", "content": "LOL, man stole my blog post hahaha
", "rawMarkdown": "LOL, [man stole my blog post](https://towardsdatascience.com/identifying-the-number-of-open-ion-channels-with-hidden-markov-models-334fab86fc85) hahaha", "votes": 3 } ] }, { "id": 975292, "postDate": "2020-08-18T08:27:28.700Z", "content": "I think this is just the tip of the iceberg.
", "rawMarkdown": "I think this is just the tip of the iceberg. ", "votes": 14, "replies": [ { "id": 975520, "postDate": "2020-08-18T10:33:33.687Z", "content": "Hopefully, it is a small iceberg then. :D
", "rawMarkdown": "Hopefully, it is a small iceberg then. :D", "votes": 1 }, { "id": 976923, "postDate": "2020-08-19T07:16:12.007Z", "content": "You are probably right, the No-Code-Only-CSV-Submission-Competitions make sharing and cheating easy - and so many found it too intriguing to not try it.
\nFraud detection competition incoming ! ;)
", "rawMarkdown": "You are probably right, the No-Code-Only-CSV-Submission-Competitions make sharing and cheating easy - and so many found it too intriguing to not try it. \n\nFraud detection competition incoming ! ;)", "votes": 2 } ] }, { "id": 974169, "postDate": "2020-08-17T19:06:12.163Z", "content": "@group16 Amazing catch. This is really important and needs immediate action.
\nThe suspected team member of yours that you've hinted, is he/she has 4/8 V100? 😂
\nI like to add one thing, in our multilingual toxic classification competition, we had a team of 5 members. At first, We competed alone but approximately 2 weeks before the deadline we're invited to join a team (4 members) around same scores.
The person who invited me was good in NLP and so we thought it would be great. After joining I found that out of four, 2 members have already gone completely silent. It was rest 3 of us. And guess what, they're from China (both). After that one has become competition expert without doing anything. After the competition, those users congratulated us when we posted our solution in the discussion. 😅
\nYour findings clarify many things to me. I always wonder that how can Joined 2/3/4/5/... days ago surpass many top kaggler (the real heroes). I don't want to repeat much, but now I can tell most of them are from China/Beijing. 😯
I hope private sharing doesn't break out like the corona epidemic :)
", "rawMarkdown": "I hope private sharing doesn't break out like the corona epidemic :)", "votes": 12, "replies": [ { "id": 973672, "postDate": "2020-08-17T12:56:57.643Z", "content": "And the world waits for the Kaggle vaccine.
", "rawMarkdown": "And the world waits for the Kaggle vaccine.", "votes": 9 }, { "id": 974775, "postDate": "2020-08-18T03:45:14.033Z", "content": "Well, this is like a type of virus and we don't have a vaccine yet…just like the coronavirus.
", "rawMarkdown": "Well, this is like a type of virus and we don't have a vaccine yet...just like the coronavirus." } ] }, { "id": 973664, "postDate": "2020-08-17T12:50:05.857Z", "content": "Anything that has monetary gains will have scammers. I don't know how much of a priority Kaggle considers this to be but I won't be surprised if there is increasing amount of unfair practises in the whole system due to historic cases of cheaters getting away with it.
\nBut certainly if this continues it will be a shame that Competitions (also) is not clean anymore and it will be curtains down for many.
", "rawMarkdown": "Anything that has monetary gains will have scammers. I don't know how much of a priority Kaggle considers this to be but I won't be surprised if there is increasing amount of unfair practises in the whole system due to historic cases of cheaters getting away with it.\n\nBut certainly if this continues it will be a shame that Competitions (also) is not clean anymore and it will be curtains down for many.", "votes": 12, "replies": [ { "id": 973671, "postDate": "2020-08-17T12:56:05.393Z", "content": "It would be really good if Kaggle comes up with a dedicated moderation team for competitions. This would lead to running of a smooth, fair competition and would help in resolving confusions created in competitions like Global Wheat Detection.
\nThe team would have the responsibilities to monitor solutions, remove non-compliant solutions, actively check for offenders and blacklist them.
\nKaggle users should also be required to verify themselves through a robust user verification system, which uses a bit more data (security experts would be better at explaining this).
", "rawMarkdown": "It would be really good if Kaggle comes up with a dedicated moderation team for competitions. This would lead to running of a smooth, fair competition and would help in resolving confusions created in competitions like Global Wheat Detection.\n\nThe team would have the responsibilities to monitor solutions, remove non-compliant solutions, actively check for offenders and blacklist them.\n\nKaggle users should also be required to verify themselves through a robust user verification system, which uses a bit more data (security experts would be better at explaining this).", "votes": 4 } ] }, { "id": 974008, "postDate": "2020-08-17T16:56:59.143Z", "content": "@divyansh22 These kind of inappropriate and hateful comments are not welcome here. I trust that you will reflect and do better.
", "rawMarkdown": "@divyansh22 These kind of inappropriate and hateful comments are not welcome here. I trust that you will reflect and do better.", "votes": 10, "replies": [ { "id": 974722, "postDate": "2020-08-18T03:03:33.593Z", "rawMarkdown": "", "votes": 1, "isDeleted": true } ] }, { "id": 973997, "postDate": "2020-08-17T16:49:04.890Z", "content": "@divyansh22 Wow always think twice before posting this kind of s**t thanks
", "rawMarkdown": "@divyansh22 Wow always think twice before posting this kind of s**t thanks", "votes": 10, "replies": [ { "id": 974716, "postDate": "2020-08-18T03:02:09.370Z", "rawMarkdown": "", "votes": 1, "isDeleted": true } ] }, { "id": 973550, "postDate": "2020-08-17T11:08:48.277Z", "content": "😲😲😲 jonney has changed name to jj-yytu and he's in the team just before me !
\nPlus, he WAS EJECTED in PANDA and in ALASKA2 (he was in silver in public and now inexist in LB I remember that because we were close).
\nThat means Kaggle knows he cheated but kept his account just because he's a MASTER ?
", "rawMarkdown": "😲😲😲 jonney has changed name to [jj-yytu](https://www.kaggle.com/swsuh2003) and he's in the team just before me ! \n\nPlus, he WAS EJECTED in [PANDA](https://www.kaggle.com/c/prostate-cancer-grade-assessment/discussion/169423#945549) and in [ALASKA2](url) (he was in silver in public and now inexist in LB I remember that because we were close).\n\nThat means Kaggle knows he cheated but kept his account just because he's a MASTER ?", "votes": 10, "replies": [ { "id": 973564, "postDate": "2020-08-17T11:19:08.463Z", "content": "Now I understand how they can use 30+ submits to catch my team with 300+ submits.
", "rawMarkdown": "Now I understand how they can use 30+ submits to catch my team with 300+ submits.", "votes": 4 }, { "id": 973634, "postDate": "2020-08-17T12:16:34.287Z", "content": "We have blocked the accounts of GMs in the past for cheating. When we have evidence of cheating, we take action. Sometimes, though, it takes time to collect and review all of the evidence.
", "rawMarkdown": "We have blocked the accounts of GMs in the past for cheating. When we have evidence of cheating, we take action. Sometimes, though, it takes time to collect and review all of the evidence. ", "votes": 26 }, { "id": 974296, "postDate": "2020-08-17T21:17:12.267Z", "content": "@vicioussong I hope you are wrong about that. It's hard to understand why Kaggle would allow someone to stay who was caught and even ejected in competitions. And then what? They see he continues to have newbies in his groups who contribute nothing, and they think anything has changed?
", "rawMarkdown": "@vicioussong I hope you are wrong about that. It's hard to understand why Kaggle would allow someone to stay who was caught and even ejected in competitions. And then what? They see he continues to have newbies in his groups who contribute nothing, and they think anything has changed?" } ] }, { "id": 973486, "postDate": "2020-08-17T10:23:48.790Z", "content": "I don't understand who pays for Kaggle medals. I have never seen any benefit of having Kaggle medals.
", "rawMarkdown": "I don't understand who pays for Kaggle medals. I have never seen any benefit of having Kaggle medals.", "votes": 11, "replies": [ { "id": 973491, "postDate": "2020-08-17T10:25:24.233Z", "content": "It seems to help a lot when looking for jobs. Especially in larger companies where the hiring is done by HR people, whose technical skills are not sufficient to see through this scam. Also @aerdem4, aren't you working for a team that recruited people based on the fact that they are kaggle grandmasters? :P
", "rawMarkdown": "It seems to help a lot when looking for jobs. Especially in larger companies where the hiring is done by HR people, whose technical skills are not sufficient to see through this scam. Also @aerdem4, aren't you working for a team that recruited people based on the fact that they are kaggle grandmasters? :P", "votes": 14 }, { "id": 973500, "postDate": "2020-08-17T10:31:41.633Z", "content": "It can be used to pad your resume and is also a good talking point in interviews if you have limited professional experience.
", "rawMarkdown": "It can be used to pad your resume and is also a good talking point in interviews if you have limited professional experience." }, { "id": 973516, "postDate": "2020-08-17T10:45:22.037Z", "content": "Unfortunately, I have opposite experience. I don't get my CV pass screening when I have Kaggle on it. And when recruiters call, they don't talk about Kaggle but work experience. They don't even know Kaggle.
\nNvidia and H2O are kind of unique in hiring Kagglers. I don't see other companies doing that. I am also against hiring people just because of their Kaggle title. You can become a grandmaster without solid knowledge in machine learning.
", "rawMarkdown": "Unfortunately, I have opposite experience. I don't get my CV pass screening when I have Kaggle on it. And when recruiters call, they don't talk about Kaggle but work experience. They don't even know Kaggle. \n\nNvidia and H2O are kind of unique in hiring Kagglers. I don't see other companies doing that. I am also against hiring people just because of their Kaggle title. You can become a grandmaster without solid knowledge in machine learning.", "votes": 14 }, { "id": 973518, "postDate": "2020-08-17T10:46:51.277Z", "content": "Btw it is actually like buying a World of Warcraft character. You may buy the best character but you won't enjoy it since your skills don't match on that competitive level.
", "rawMarkdown": "Btw it is actually like buying a World of Warcraft character. You may buy the best character but you won't enjoy it since your skills don't match on that competitive level.", "votes": 19 }, { "id": 973526, "postDate": "2020-08-17T10:50:17.940Z", "content": "I remember that in this Melanoma competition, at the very starting, one guy from Japan commented that \"Please don't submit high scoring subs, otherwise I won't get any medal, in Japan, the companies prefer Kaggle medals.\"
", "rawMarkdown": "I remember that in this Melanoma competition, at the very starting, one guy from Japan commented that \"Please don't submit high scoring subs, otherwise I won't get any medal, in Japan, the companies prefer Kaggle medals.\"" }, { "id": 973533, "postDate": "2020-08-17T10:54:18.160Z", "content": "Some companies actually list Kaggle experience as a bonus in their listing. Twice interviewers have asked me about specific comps I attended and the techniques I used.
\nThat said, I doubt the medals worth as much as much as actual professional experience.
", "rawMarkdown": "Some companies actually list Kaggle experience as a bonus in their listing. Twice interviewers have asked me about specific comps I attended and the techniques I used.\n\nThat said, I doubt the medals worth as much as much as actual professional experience." }, { "id": 973540, "postDate": "2020-08-17T10:57:39.590Z", "content": "I believe a Kaggle competition may worth even 1 year of work experience depending on the work and the competition. Especially in Europe, work usually progresses slow. But on the other hand, not all competitions are very fruitful, some of them are lottery.
", "rawMarkdown": "I believe a Kaggle competition may worth even 1 year of work experience depending on the work and the competition. Especially in Europe, work usually progresses slow. But on the other hand, not all competitions are very fruitful, some of them are lottery.", "votes": 2 }, { "id": 973544, "postDate": "2020-08-17T11:01:29.847Z", "content": "Yeah, actually that's good that they focus on Kaggle competitions and ask the candidates about how they approached the competition because a competition here is about 3 months long, so it provides a interviewer the opportunity to ask many questions and ask thoroughly. Also, it is a huge boost to Kaggle's Image.
\nBut, if a company knows about such events, it will definitely malign the image of the community.
\nWhat we need now are some serious changes, so that no outsider can exploit the community and its culture.
", "rawMarkdown": "Yeah, actually that's good that they focus on Kaggle competitions and ask the candidates about how they approached the competition because a competition here is about 3 months long, so it provides a interviewer the opportunity to ask many questions and ask thoroughly. Also, it is a huge boost to Kaggle's Image.\n\nBut, if a company knows about such events, it will definitely malign the image of the community.\n\nWhat we need now are some serious changes, so that no outsider can exploit the community and its culture." }, { "id": 973548, "postDate": "2020-08-17T11:06:54.873Z", "content": "I mean I don't think any sensible person is going to hire you solely for your medal lol. It's just a way to get pass the non-technical HR. You still need to clear the technical interview regardless so I don't think those people that bought medal is going to benefit much from it.
", "rawMarkdown": "I mean I don't think any sensible person is going to hire you solely for your medal lol. It's just a way to get pass the non-technical HR. You still need to clear the technical interview regardless so I don't think those people that bought medal is going to benefit much from it.", "votes": 3 }, { "id": 973570, "postDate": "2020-08-17T11:24:25.673Z", "content": "Yeah, that's totally true!
", "rawMarkdown": "Yeah, that's totally true!" }, { "id": 973597, "postDate": "2020-08-17T11:45:19.280Z", "content": "If I remember correctly, I saw a post about the job market in China being very competitive. People invest lots of money to find an entry level job.
", "rawMarkdown": "If I remember correctly, I saw a post about the job market in China being very competitive. People invest lots of money to find an entry level job." }, { "id": 973750, "postDate": "2020-08-17T13:49:24.693Z", "content": "I think it's competitive everywhere for entry level. New grads are a dime a dozen but companies (understandably) only want seniors.
", "rawMarkdown": "I think it's competitive everywhere for entry level. New grads are a dime a dozen but companies (understandably) only want seniors." } ] }, { "id": 974004, "postDate": "2020-08-17T16:54:13.117Z", "content": "@divyansh22 Don't vent your anger on a specific country. Chinese competitors are not responsible for this small group of people. Your comment is totally racist. What has Kaggle to do COVID?
\n(Kaggle discussions seem to be broken. I can't reply to comments)
", "rawMarkdown": "@divyansh22 Don't vent your anger on a specific country. Chinese competitors are not responsible for this small group of people. Your comment is totally racist. What has Kaggle to do COVID?\n\n(Kaggle discussions seem to be broken. I can't reply to comments)", "votes": 8, "replies": [ { "id": 974717, "postDate": "2020-08-18T03:02:44.700Z", "rawMarkdown": "", "votes": 2, "isDeleted": true }, { "id": 974999, "postDate": "2020-08-18T05:35:31.603Z", "content": "I'm glad that you accepted your fault. Hope you have a great day :)
", "rawMarkdown": "I'm glad that you accepted your fault. Hope you have a great day :)", "votes": 1 } ] }, { "id": 974656, "postDate": "2020-08-18T02:19:42.583Z", "content": "I think it's time to make at least ID and face verification as in EdX, at least. This will prevent such issues not 100%, but mostly.
\nAlso it will help good Kagglers not to team up someone wrong who can cheat and crash the work. Happened to my team at least once (proven one), which led to loss of TOP3% silver and 2.5 month of work.
", "rawMarkdown": "I think it's time to make at least ID and face verification as in EdX, at least. This will prevent such issues not 100%, but mostly.\n\nAlso it will help good Kagglers not to team up someone wrong who can cheat and crash the work. Happened to my team at least once (proven one), which led to loss of TOP3% silver and 2.5 month of work. ", "votes": 7, "replies": [ { "id": 975526, "postDate": "2020-08-18T10:37:40.623Z", "content": "That's a great idea: requiring some level of id would be a deterrent to a lot of \"fake\" accounts. I guess the technical implementation should take some point though. Or maybe it is something that Kaggle is already working on, we will see :)
", "rawMarkdown": "That's a great idea: requiring some level of id would be a deterrent to a lot of \"fake\" accounts. I guess the technical implementation should take some point though. Or maybe it is something that Kaggle is already working on, we will see :)" } ] }, { "id": 973783, "postDate": "2020-08-17T14:09:54.873Z", "content": "What the hell is this???????? They are totally undermining the efforts of many Kagglers. Kaggle should take strong action and ban all those users who got any single medal using such platform.
", "rawMarkdown": "What the hell is this???????? They are totally undermining the efforts of many Kagglers. Kaggle should take strong action and ban all those users who got any single medal using such platform.", "votes": 7 }, { "id": 974246, "postDate": "2020-08-17T20:31:26.850Z", "content": "Amazing investigative work! That's very bad for the Kaggle community and makes the ranking a little less meaningful, which is sad. ☹️
\nThat being said, nothing beats the joy of learning something new and exploring new ideas, so I guess these that pay for medals won't get that. 😄
\nFinally, if I may ask, do you know why you were contacted? Did it happen after you teamed up with the suspicious team-mate or was that just a coincidence?
\nAgain, thanks for the post!
", "rawMarkdown": "Amazing investigative work! That's very bad for the Kaggle community and makes the ranking a little less meaningful, which is sad. ☹️\n\nThat being said, nothing beats the joy of learning something new and exploring new ideas, so I guess these that pay for medals won't get that. 😄\n\nFinally, if I may ask, do you know why you were contacted? Did it happen after you teamed up with the suspicious team-mate or was that just a coincidence?\n\nAgain, thanks for the post!", "votes": 8, "replies": [ { "id": 974248, "postDate": "2020-08-17T20:33:42.540Z", "content": "Probably due to my previous post :)
", "rawMarkdown": "[Probably due to my previous post :)](https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/174635)", "votes": 1, "replies": [ { "id": 974260, "postDate": "2020-08-17T20:48:46.097Z", "content": "It wouldn't be necessarily, a bad idea, to offer some sort of amnesty. Those who come forward, confess, are not sanctioned and not doxed. Perhaps their medals are stripped…….I mean, there are consequences, but if they are a real person, not a bot/mule/farm account, then they can be given a clean slate going forward and they would be doing us all a big help, by turning in as much info as possible.
", "rawMarkdown": "It wouldn't be necessarily, a bad idea, to offer some sort of amnesty. Those who come forward, confess, are not sanctioned and not doxed. Perhaps their medals are stripped.......I mean, there are consequences, but if they are a real person, not a bot/mule/farm account, then they can be given a clean slate going forward and they would be doing us all a big help, by turning in as much info as possible. \n", "votes": 2 } ] }, { "id": 974252, "postDate": "2020-08-17T20:37:02.993Z", "content": "Will give it a look, cheers!
", "rawMarkdown": "Will give it a look, cheers!" } ] }, { "id": 974221, "postDate": "2020-08-17T20:00:14.713Z", "content": "The value of things is determined by how easy or hard is to get them. If it will be possible to simply buy Kaggle medals, their value will decrease. So we all, including Kaggle and competitors, are interested that cheaters are revealed and their medals removed.
", "rawMarkdown": "The value of things is determined by how easy or hard is to get them. If it will be possible to simply buy Kaggle medals, their value will decrease. So we all, including Kaggle and competitors, are interested that cheaters are revealed and their medals removed.", "votes": 8 }, { "id": 974115, "postDate": "2020-08-17T18:16:27.550Z", "content": "@inversion plenty of evidence here and also here https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/174635#973667 .
\nedu.angtk.com took their website down as @group16 , I and others were challenging them last night. You see the main representative https://www.kaggle.com/lifeifei38 , who apparently is part of the staff for this medal farm, has an account a few months old with no activity. You can't let them get away with these medals in Wheat and Melanoma……….the lid is blown open on this group, and they have been doing this apparently for at least 2 years, but it's blatant, \"in your face\" now. I mean just look at the groups with user @yikououbao . Thats pure medal farming and yikououbao has taken real medals, and prevented many others from getting them.
", "rawMarkdown": "@inversion plenty of evidence here and also here https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/174635#973667 .\n\nedu.angtk.com took their website down as @group16 , I and others were challenging them last night. You see the main representative https://www.kaggle.com/lifeifei38 , who apparently is part of the staff for this medal farm, has an account a few months old with no activity. You can't let them get away with these medals in Wheat and Melanoma..........the lid is blown open on this group, and they have been doing this apparently for at least 2 years, but it's blatant, \"in your face\" now. I mean just look at the groups with user @yikououbao . Thats pure medal farming and yikououbao has taken real medals, and prevented many others from getting them. ", "votes": 8 }, { "id": 973647, "postDate": "2020-08-17T12:27:53.707Z", "content": "I think the danger of this situation cannot be stressed enough: It threatens to undermine the community based on the open-source premise which makes it fair for all to compete, regardless of sociodemographic background. Open, and transparent collaboration strengthens the community by building trust in the methods and helping to push the cutting edge.
\nCompanies / methods like this are toxic and damage the progress made by the wider open source community and the ideals they champion. Taking such large sums of money off people who may not know better is a low-grade business tactic - shame on them.
\n$500 is a lot of money even for someone from the US/EU! For comparison, High-caliber ML courses vary from free to tens of dollars…
", "rawMarkdown": "I think the danger of this situation cannot be stressed enough: It threatens to undermine the community based on the open-source premise which makes it fair for all to compete, regardless of sociodemographic background. Open, and transparent collaboration strengthens the community by building trust in the methods and helping to push the cutting edge.\n\nCompanies / methods like this are toxic and damage the progress made by the wider open source community and the ideals they champion. Taking such large sums of money off people who may not know better is a low-grade business tactic - shame on them.\n\n $500 is a lot of money even for someone from the US/EU! For comparison, High-caliber ML courses vary from free to tens of dollars...", "votes": 8 }, { "id": 973492, "postDate": "2020-08-17T10:28:52.630Z", "content": "The account you mention looks highly likely to be a medal farm that tricks its way onto high-scoring teams. I expect the photo was chosen to increase credibility and to influence people.
", "rawMarkdown": "The account you mention looks highly likely to be a medal farm that tricks its way onto high-scoring teams. I expect the photo was chosen to increase credibility and to influence people.", "votes": 8, "replies": [ { "id": 973499, "postDate": "2020-08-17T10:31:23.530Z", "content": "My thoughts exactly. I just hope that it is only that and that the account is in no way involved with the private sharing. I'll gladly help the people from Kaggle in any way possible to further clear this out.
", "rawMarkdown": "My thoughts exactly. I just hope that it is only that and that the account is in no way involved with the private sharing. I'll gladly help the people from Kaggle in any way possible to further clear this out.", "votes": 2 }, { "id": 974749, "postDate": "2020-08-18T03:25:25.193Z", "content": "Very typical \"successful customer story\" tricks in the marketing domain.
", "rawMarkdown": "Very typical \"successful customer story\" tricks in the marketing domain." } ] }, { "id": 974140, "postDate": "2020-08-17T18:38:39.130Z", "content": "As a side note, and for what it's worth, I feel like it would be perfectly fine for the private LB validation to take more time in general. This doesn't have to be an instantaneous switch; especially if the additional time helps to make sure that all the final placings are legit. As long as the time scales are clearly communicated, some delay might even help in keeping more people interested in post-submission deadline discussions and solution sharing.
", "rawMarkdown": "As a side note, and for what it's worth, I feel like it would be perfectly fine for the private LB validation to take more time in general. This doesn't have to be an instantaneous switch; especially if the additional time helps to make sure that all the final placings are legit. As long as the time scales are clearly communicated, some delay might even help in keeping more people interested in post-submission deadline discussions and solution sharing.", "votes": 5, "replies": [ { "id": 974148, "postDate": "2020-08-17T18:44:09.203Z", "content": "It's OK to take SOME more time instead of instant switching, but we don't need another ASHRAE.
", "rawMarkdown": "It's OK to take SOME more time instead of instant switching, but we don't need another ASHRAE.", "votes": 1 } ] }, { "id": 973730, "postDate": "2020-08-17T13:34:57.720Z", "content": "First of all, you did a great job finding this.
\nSecond, Kaggle should take necessary steps against these users. It discourages beginners and disrupts the whole sharing and learning process among the community.
In one of their blog post on ion switching competition: https://zhuanlan.zhihu.com/p/150559937
\nAn image caption as: \"top 15% can be reached using our early-stage solution\"
\n![\"](\"https://picb.zhimg.com/80/v2-c97cbd4dc6028ef0718a11d2c8bbd7f3_720w.jpg\")
It's just blatant private sharing?
", "rawMarkdown": "In one of their blog post on ion switching competition: https://zhuanlan.zhihu.com/p/150559937 \n\nAn image caption as: \"top 15% can be reached using our early-stage solution\"\n\n\nIt's just blatant private sharing?", "votes": 5, "replies": [ { "id": 974315, "postDate": "2020-08-17T22:01:20.370Z", "content": "that link is now 404. They are trying to cover all the tracks.
", "rawMarkdown": "that link is now 404. They are trying to cover all the tracks. " } ] }, { "id": 973592, "postDate": "2020-08-17T11:41:36.763Z", "content": "This is becoming Kaggle CSI, indeed it is very hard to catch this things in a systematic way with concrete proof.
\nAnyway, it is interesting that people think that Kaggle medals will give them jobs, just by having them, any serious company will make sure you know about things, a lot of companies don't even care about Kaggle.
", "rawMarkdown": "This is becoming Kaggle CSI, indeed it is very hard to catch this things in a systematic way with concrete proof.\n\nAnyway, it is interesting that people think that Kaggle medals will give them jobs, just by having them, any serious company will make sure you know about things, a lot of companies don't even care about Kaggle.", "votes": 5, "replies": [ { "id": 973626, "postDate": "2020-08-17T12:11:13.247Z", "content": "Yes, exactly how the certification works. You don't just get jobs based on certification, you need to show what skills you learnt. Sad that some people take this route.
", "rawMarkdown": "Yes, exactly how the certification works. You don't just get jobs based on certification, you need to show what skills you learnt. Sad that some people take this route.", "votes": 1 }, { "id": 974761, "postDate": "2020-08-18T03:33:33.887Z", "content": "I think medals are not only used for job hunting. It's more applicable to university applications or something like that.
", "rawMarkdown": "I think medals are not only used for job hunting. It's more applicable to university applications or something like that." } ] }, { "id": 976179, "postDate": "2020-08-18T17:38:09.773Z", "content": "Jumped up 5 places, cheaters getting rid of?
", "rawMarkdown": "Jumped up 5 places, cheaters getting rid of?", "votes": 6, "replies": [ { "id": 976183, "postDate": "2020-08-18T17:41:55.480Z", "content": "You were already at the top bro, you won't have this much fun that I'm having, I jumped from 687 to 625 lol! 😄😄
\nSeems like a lot of it is about to come. Happy shaking! :)
", "rawMarkdown": "You were already at the top bro, you won't have this much fun that I'm having, I jumped from 687 to 625 lol! 😄😄\n\nSeems like a lot of it is about to come. Happy shaking! :)" }, { "id": 976185, "postDate": "2020-08-18T17:43:51.427Z", "content": "Actually I was happy at 23, now I feel like I just miss solo gold at edge 😄
", "rawMarkdown": "Actually I was happy at 23, now I feel like I just miss solo gold at edge 😄", "votes": 6 }, { "id": 976188, "postDate": "2020-08-18T17:45:19.920Z", "content": "Yeah, this sadness is of another level. 😄
", "rawMarkdown": "Yeah, this sadness is of another level. 😄", "votes": 2 }, { "id": 976672, "postDate": "2020-08-19T03:03:11.867Z", "content": "After one night, I found I'm a goal keeper of silver zone…
", "rawMarkdown": "After one night, I found I'm a goal keeper of silver zone...", "votes": 1 }, { "id": 976675, "postDate": "2020-08-19T03:06:05.627Z", "content": "glad to see you get what you deserve
", "rawMarkdown": "glad to see you get what you deserve", "votes": 1 } ] }, { "id": 974310, "postDate": "2020-08-17T21:34:14.223Z", "content": "It seems the LB is being screwed up again….
\nI see many novice in solo and teams who jumped lately in range 70th to 80th with very few submissions.
\nFor instance this one just suscribed 10 days ago.
", "rawMarkdown": "It seems the LB is being screwed up again....\n\nI see many novice in solo and teams who jumped lately in range 70th to 80th with very few submissions. \n\nFor instance [this one](https://www.kaggle.com/catting) just suscribed 10 days ago. ", "votes": 4, "replies": [ { "id": 974346, "postDate": "2020-08-17T23:12:44.240Z", "content": "I think kaggle should make facial verification while a new user wants to create an account in kaggle. Or any sort of Identification number (for ex. NID) that is very unique per user or several stages of verification to create an account in kaggle.
\nSuch 10 days heroes (LOL) are in every competition. And guess what, most of the time I've noticed that, they picked cartoonish profile pictures that indicates (somewhat) from same culture people.
I like your idea. They could mail code to your physical address. Any change of address may even require a new code to be mailed. A picture using a detection algorithm to ensure its a live pic that would need to be public otherwise \"who cares?\" I do believe nothing is perfect, but multiple stages of control is key.
\nIn the end, if they had a model that looked at things like:
\nDate Joined
\nProfile pic (is it a default goose?)
\nPost/Comment/Like activity
\nSubmissions
\nTeam Mates (are they also newbs)
\nHow you climb the ladder, is it at the same time as other suspects
\nEven the country you are from if that is a positive factor
Then you just kill the accounts. Just like Credit Card companies cut you off. You have to call in, and explain any activity to get re-activated. Sure you have some False Positives, but you know you are doing a good job of catching people.
", "rawMarkdown": "I like your idea. They could mail code to your physical address. Any change of address may even require a new code to be mailed. A picture using a detection algorithm to ensure its a live pic that would need to be public otherwise \"who cares?\" I do believe nothing is perfect, but multiple stages of control is key.\n\nIn the end, if they had a model that looked at things like:\nDate Joined\nProfile pic (is it a default goose?)\nPost/Comment/Like activity\nSubmissions\nTeam Mates (are they also newbs)\nHow you climb the ladder, is it at the same time as other suspects\nEven the country you are from if that is a positive factor\n\nThen you just kill the accounts. Just like Credit Card companies cut you off. You have to call in, and explain any activity to get re-activated. Sure you have some False Positives, but you know you are doing a good job of catching people. ", "votes": 2 }, { "id": 974677, "postDate": "2020-08-18T02:29:12.017Z", "content": "Agreed. However, I don't get one thing, kaggle is now google property, why they do not ensure authenticity in the first place!
", "rawMarkdown": "Agreed. However, I don't get one thing, kaggle is now google property, why they do not ensure authenticity in the first place! " }, { "id": 974686, "postDate": "2020-08-18T02:35:05.120Z", "content": "One thing I like in Stack-Overflow is newbie restriction. If I'm just open my acc on SO, I can't downvote, can't make comment…., etc, and in order to do that I need a level of reputation first (I guess 15). I think kaggle should also deploy similar strategy, it surely can come helpful.
", "rawMarkdown": "One thing I like in Stack-Overflow is **newbie restriction**. If I'm just open my acc on SO, I can't downvote, can't make comment...., etc, and in order to do that I need a level of **reputation** first (I guess 15). I think kaggle should also deploy similar strategy, it surely can come helpful.", "votes": 2 }, { "id": 974689, "postDate": "2020-08-18T02:38:17.430Z", "content": "well you know gmail is full of duplicate and fake accounts, I am sure. But yes, for something like this you would expect more. Maybe they implement more after this.
", "rawMarkdown": "well you know gmail is full of duplicate and fake accounts, I am sure. But yes, for something like this you would expect more. Maybe they implement more after this. ", "votes": 1 }, { "id": 974696, "postDate": "2020-08-18T02:39:15.433Z", "content": "I guess maybe featured competitions should be open only for users who are expert or greater, or maybe contributor with some qualification like, at least 2 notebooks published or so, that will ensure authenticity too!
", "rawMarkdown": "I guess maybe featured competitions should be open only for users who are expert or greater, or maybe contributor with some qualification like, at least 2 notebooks published or so, that will ensure authenticity too!", "votes": 1 } ] } ] }, { "id": 973509, "postDate": "2020-08-17T10:38:26.523Z", "content": "That's seriously Awful, I feel pity for those who ask for upvotes on LinkedIn after seeing this, those guys deserve way more respect than these people.
", "rawMarkdown": "That's seriously Awful, I feel pity for those who ask for upvotes on LinkedIn after seeing this, those guys deserve way more respect than these people. ", "votes": 4 }, { "id": 973493, "postDate": "2020-08-17T10:29:32.680Z", "content": "You did a great job finding this. I haven't participated in this competition, but I have been following what the community has been doing. I honestly was feared seeing how an open platform is being exploited.
\nThe last 24 hours were a mess. Private sharing and the $1000 course.
\nIt's great to see people like you and @brianfeeny bringing up some conclusive evidence. I hope the Kaggle team works on this issue, please don't let this happen in the future competitions.
\nPeople work hard to get themselves in the top, it's disheartening to see their hardwork wash away by some malicious people.
\nWhat the Kaggle team can do for the future:
\nThanks for @group16 & other Kagglers to dig this out. Definitely support for necessary actions to clean our community. Besides such cheating for medals, I'm also speechless about \"cheating for upvotes via folked notebooks\" activities in recent competitions.
\nI'm thrilled to see so many GM's are in this competition and look forward to learning from their high-quality kernels/notebooks. E.g. Chris did a fantastic job and I learnt so much from the series of notebooks. But I feel uncomfortable to see people \"copy & paste\" others' original work, reproduce some \"high-score kernels\" with same method, input and logic, and ask for upvotes presumptuously. In the Wheat detection, I've seen such activity many times. Sometimes the post is not even related to the competition and the author would just scam everywhere and reply \"Great work! Check and upvote my EDA/notebook here!\"….OK, a notebook grandmaster/master/expert should be also a great proof of efforts, totally agree with that. But earning it in this way has no difference from antag.ai people actually. I know this is some \"self-marketing\" tricks, but getting a Stanford online certificate does not make real Stanford alumni.
\nAfter going through the comments, what drove me a little bit more uncomfortable is the \"hateful\" stereotype towards a specific country. Now the issue becomes a counterpart of coronavirus in Kaggle. When the pandemic firstly starts, some people call it \"China virus\". But should we call it \"US virus\" now? What matters is the \"vaccine\", not the name for the virus…
", "rawMarkdown": "Thanks for @group16 & other Kagglers to dig this out. Definitely support for necessary actions to clean our community. Besides such cheating for medals, I'm also speechless about \"cheating for upvotes via folked notebooks\" activities in recent competitions. \n\nI'm thrilled to see so many GM's are in this competition and look forward to learning from their high-quality kernels/notebooks. E.g. Chris did a fantastic job and I learnt so much from the series of notebooks. But I feel uncomfortable to see people \"copy & paste\" others' original work, reproduce some \"high-score kernels\" with same method, input and logic, and ask for upvotes presumptuously. In the Wheat detection, I've seen such activity many times. Sometimes the post is not even related to the competition and the author would just scam everywhere and reply \"Great work! Check and upvote my EDA/notebook here!\"....OK, a notebook grandmaster/master/expert should be also a great proof of efforts, totally agree with that. But earning it in this way has no difference from antag.ai people actually. I know this is some \"self-marketing\" tricks, but getting a Stanford online certificate does not make real Stanford alumni.\n\nAfter going through the comments, what drove me a little bit more uncomfortable is the \"hateful\" stereotype towards a specific country. Now the issue becomes a counterpart of coronavirus in Kaggle. When the pandemic firstly starts, some people call it \"China virus\". But should we call it \"US virus\" now? What matters is the \"vaccine\", not the name for the virus...", "votes": 3 }, { "id": 979044, "postDate": "2020-08-20T15:25:55.123Z", "content": "Dear fellow Sherlocks
\nthose two accounts seem very much to belong to just one person:
\nhttps://www.kaggle.com/jiuntianrobot
\nhttps://www.kaggle.com/jiuntian
They follow each other & the robot account only handed in solutions on those days where jiuntian was handing in solutions.
\nToo bad for our colleague: The robot chose the final submissions more wisely ( = higher rank ;))
https://www.kaggle.com/tempusername1234 <-- looks also suspicious, but I cant link him yet to our magnificent 7 medal-buyers.
\nhttps://www.kaggle.com/qwert217 <-- same
", "rawMarkdown": "Dear fellow Sherlocks\n\nthose two accounts seem very much to belong to just one person:\nhttps://www.kaggle.com/jiuntianrobot\nhttps://www.kaggle.com/jiuntian\n\nThey follow each other & the robot account only handed in solutions on those days where jiuntian was handing in solutions.\nToo bad for our colleague: The robot chose the final submissions more wisely ( = higher rank ;))\n\n\nhttps://www.kaggle.com/tempusername1234 <-- looks also suspicious, but I cant link him yet to our magnificent 7 medal-buyers.\n\nhttps://www.kaggle.com/qwert217 <-- same", "votes": 3, "replies": [ { "id": 979152, "postDate": "2020-08-20T16:55:18.093Z", "content": "Oh definitely, @inversion did you see this one?
", "rawMarkdown": "Oh definitely, @inversion did you see this one?\n", "votes": 2 }, { "id": 979205, "postDate": "2020-08-20T17:33:03.330Z", "content": "@inversion, @philculliton, this account i.e. @yikououbao reported as suspicious and linked to edu.angtk.com is still there as well. She/he have real medals which should be revoked and the LB of those competitions amended accordingly.
", "rawMarkdown": "@inversion, @philculliton, this account i.e. @yikououbao reported as suspicious and linked to edu.angtk.com is still there as well. She/he have real medals which should be revoked and the LB of those competitions amended accordingly.", "votes": 4 }, { "id": 979221, "postDate": "2020-08-20T17:43:45.963Z", "content": "Agreed! Its one of the ones that has caused the most stir, yet still exists. I have plans to compile some of my evidence and submit, I didn't think it would have to come to that since there is so much out there in outrage on that team.
", "rawMarkdown": "Agreed! Its one of the ones that has caused the most stir, yet still exists. I have plans to compile some of my evidence and submit, I didn't think it would have to come to that since there is so much out there in outrage on that team.", "votes": 1 }, { "id": 979225, "postDate": "2020-08-20T17:44:45.923Z", "content": "And it's not just @yikououbao its everyone she has teamed with, all suspicious
", "rawMarkdown": "And it's not just @yikououbao its everyone she has teamed with, all suspicious", "votes": 4 }, { "id": 979235, "postDate": "2020-08-20T17:51:08.367Z", "content": "Noted. Thx!
", "rawMarkdown": "Noted. Thx!", "votes": 4 }, { "id": 979244, "postDate": "2020-08-20T17:56:48.150Z", "content": "I agree @brianfeeny. I posted in another thread that all the noobs he/she teamed up with are one timers and only one i.e. @yauwaiyim is a repeat team member. Note that @yauwaiyim teamed up with the same people for all 4 competitions he/she participated in.
", "rawMarkdown": "I agree @brianfeeny. I posted in another thread that all the noobs he/she teamed up with are one timers and only one i.e. @yauwaiyim is a repeat team member. Note that @yauwaiyim teamed up with the same people for all 4 competitions he/she participated in.", "votes": 1 }, { "id": 979282, "postDate": "2020-08-20T18:29:01.877Z", "content": "Thank you for all of your reports! I want to emphasize that we are only able to take specific action on those teams where evidence is conclusive. Simply having a pattern of teaming with novice accounts may raise suspicion, but is an insufficient qualification for removal. Those who are suspicious will remain very close to our radar, but please be aware that if you do not witness a removal of every team you've identified in the multiple threads attempting to call out possible offenders, that is why. But trust that we are taking this very seriously.
", "rawMarkdown": "Thank you for all of your reports! I want to emphasize that we are only able to take specific action on those teams where evidence is conclusive. Simply having a pattern of teaming with novice accounts may raise suspicion, but is an insufficient qualification for removal. Those who are suspicious will remain very close to our radar, but please be aware that if you do not witness a removal of every team you've identified in the multiple threads attempting to call out possible offenders, that is why. But trust that we are taking this **very seriously**.", "votes": 5 }, { "id": 979328, "postDate": "2020-08-20T19:04:43.237Z", "content": "I would just like to comment, that the industry standard in fraud, is not \"beyond a reasonable doubt\". I mean, it is, for final adjudication, but it is not for flagging. At the extreme, say someone joins, submits 4 times, and is in the medal zone. Do you have any proof they did anything wrong? No. But this is a Data Science competition, and in Science, we rely on statistics, and we can all agree with almost no doubt that something wrong is happening. If you are going to be a Data Science competition site, you must embrace the principles of Science.
\nThanks for all the cleaning up, but I think you all are making a huge mistake if you don't go after yikououbao and the whole network of teammates (in current and past competitions), some of which have been found to be fraudulent and deleted. Fraudsters team with other fraudsters. Birds of a Feather. The exception is those that try to be \"sleepers\" where they latch on to a Master/GM team and just take advantage……….But when you are actually stealing code solutions, a la angtk.com, you can't have legitimate people in your group, that won't work.
\nWhen a credit card flags your account, it's not because they have \"beyond a reasonable doubt\". It's because they have a suspicion. No one can argue yikououbao is not EXTREMELY suspicious. That is where you intervene, ask questions, look deeper, and make a determination. People depend on you all for that. Otherwise, you allow really clever people to slip through all the cracks and that is not fair to anyone. I guess you may feel otherwise, but @juliaelliott there are not legitimate people on Kaggle that join, never post, never interact in any way, partner with only newbs, and takedown medals.
", "rawMarkdown": "I would just like to comment, that the industry standard in fraud, is not \"beyond a reasonable doubt\". I mean, it is, for final adjudication, but it is not for flagging. At the extreme, say someone joins, submits 4 times, and is in the medal zone. Do you have any proof they did anything wrong? No. But this is a Data Science competition, and in Science, we rely on statistics, and we can all agree with almost no doubt that something wrong is happening. If you are going to be a Data Science competition site, you must embrace the principles of Science. \n\nThanks for all the cleaning up, but I think you all are making a huge mistake if you don't go after yikououbao and the whole network of teammates (in current and past competitions), some of which have been found to be fraudulent and deleted. Fraudsters team with other fraudsters. Birds of a Feather. The exception is those that try to be \"sleepers\" where they latch on to a Master/GM team and just take advantage..........But when you are actually stealing code solutions, a la angtk.com, you can't have legitimate people in your group, that won't work.\n\nWhen a credit card flags your account, it's not because they have \"beyond a reasonable doubt\". It's because they have a suspicion. No one can argue yikououbao is not EXTREMELY suspicious. That is where you intervene, ask questions, look deeper, and make a determination. People depend on you all for that. Otherwise, you allow really clever people to slip through all the cracks and that is not fair to anyone. I guess you may feel otherwise, but @juliaelliott there are not legitimate people on Kaggle that join, never post, never interact in any way, partner with only newbs, and takedown medals. ", "votes": 4 } ] }, { "id": 975476, "postDate": "2020-08-18T10:05:41.140Z", "content": "Shocking that people are actually buying and selling kaggle positions! Although I guess that's not too surprising given how they do have value. Definitely agree with what others have said about Kaggle requiring ID verification like Coursera or EdX.
", "rawMarkdown": "Shocking that people are actually buying and selling kaggle positions! Although I guess that's not too surprising given how they do have value. Definitely agree with what others have said about Kaggle requiring ID verification like Coursera or EdX.", "votes": 3 }, { "id": 975122, "postDate": "2020-08-18T06:53:18.167Z", "content": "very bad.. but all those have serious learning intent remain unmoved by such medal based temptations and they continue working hard ,take quidances from expert and keep moving..
", "rawMarkdown": "very bad.. but all those have serious learning intent remain unmoved by such medal based temptations and they continue working hard ,take quidances from expert and keep moving..", "votes": 3 }, { "id": 974661, "postDate": "2020-08-18T02:21:49.850Z", "content": "@group16 amazing job
\nHow you found this? The difficulty to search and detect suspicious stuff from this part of the world is alphabet. It complicates the search. But I see this brilliant work and appreciate that. As more such job done, as better Kaggle will be for future participants. I hope.
He has been contacted by someone so I guess he started from there and branched to other findings. @group16 do you speak Mandarin or did you use Google translate or some variant?
", "rawMarkdown": "He has been contacted by someone so I guess he started from there and branched to other findings. @group16 do you speak Mandarin or did you use Google translate or some variant?" }, { "id": 975529, "postDate": "2020-08-18T10:41:41.087Z", "content": "我當然會說普通話
\nNo, my source translated everything for me :)
", "rawMarkdown": "我當然會說普通話\n\nNo, my source translated everything for me :)", "votes": 3 }, { "id": 975541, "postDate": "2020-08-18T10:48:45.797Z", "content": "Found this great website (with audio as well): https://www.trainchinese.com/v2/wordDetails.php?rAp=0&wordId=37859&tcLanguage=en
\nI am a bit disappointed, I thought you were a Mandarin GrandMaster. :(
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F172860%2Fed347f05c311cdbdefc0e48c619b853b%2Fmandarin_text.png?generation=1597747723738602&alt=media\")
Haha sorry to have let you down. I'm still far away from ever becoming a mandarin grandmaster. Most of the text was also embedded into images. I wouldn't even know how to start with typing out these characters in google translate…
", "rawMarkdown": "Haha sorry to have let you down. I'm still far away from ever becoming a mandarin grandmaster. Most of the text was also embedded into images. I wouldn't even know how to start with typing out these characters in google translate...", "votes": 3 }, { "id": 975559, "postDate": "2020-08-18T10:57:48.803Z", "content": "", "rawMarkdown": "https://translate.yandex.com/", "votes": 1 } ] }, { "id": 974100, "postDate": "2020-08-17T18:12:54.490Z", "content": "Rather than delete cheaters, why not keep them in a public hall of shame. That way if they're discovered and verified as being cheaters, it can always be cross-referenced both now and in the future.
", "rawMarkdown": "Rather than delete cheaters, why not keep them in a public hall of shame. That way if they're discovered and verified as being cheaters, it can always be cross-referenced both now and in the future.", "votes": 3, "replies": [ { "id": 974145, "postDate": "2020-08-17T18:42:06.007Z", "content": "Public hall of shame is old idea. Old but gold.
", "rawMarkdown": "Public hall of shame is old idea. Old but gold.", "votes": 2 }, { "id": 974157, "postDate": "2020-08-17T18:55:06.233Z", "content": "Reputation should work both ways. Also, if this is for employment, then wouldn't it be nice to see if what your prospect is touting is valid or not? If cheaters are using Kaggle to falsify their resumes, it shouldn't be swept under a rug.
", "rawMarkdown": "Reputation should work both ways. Also, if this is for employment, then wouldn't it be nice to see if what your prospect is touting is valid or not? If cheaters are using Kaggle to falsify their resumes, it shouldn't be swept under a rug. ", "votes": 2 }, { "id": 974166, "postDate": "2020-08-17T19:04:00.750Z", "content": "Some online games secretly ban their cheating players in a way that they won't even notice. They don't tell them they are banned but they secretly put them into the cheater pool. They always matched with other cheaters if they are in that pool. They do this in order to prevent them to make new accounts. I wish this could be applicable to Kaggle, but it is not.
", "rawMarkdown": "Some online games secretly ban their cheating players in a way that they won't even notice. They don't tell them they are banned but they secretly put them into the cheater pool. They always matched with other cheaters if they are in that pool. They do this in order to prevent them to make new accounts. I wish this could be applicable to Kaggle, but it is not.\n ", "votes": 1 }, { "id": 974217, "postDate": "2020-08-17T19:52:09.637Z", "content": "Oh I mean, I think it'd be great for it to be as public as possible. If you're bragging that you're a medalist, then you need to say what your account is on the resume - then the employer should be able to cross check. If you're willing to lie about a Kaggle medal, why should you be trusted as an employee? Likewise, if you're a student, it should be seen as academic misconduct.
", "rawMarkdown": "Oh I mean, I think it'd be great for it to be as public as possible. If you're bragging that you're a medalist, then you need to say what your account is on the resume - then the employer should be able to cross check. If you're willing to lie about a Kaggle medal, why should you be trusted as an employee? Likewise, if you're a student, it should be seen as academic misconduct." } ] }, { "id": 974099, "postDate": "2020-08-17T18:12:29.520Z", "content": "Interesting @group16. Thanks for investigation and sharing.
\nKaggle should step up to improve/keep its reputation.
I don't think it is possible for Kaggle team to check every single entry it will take forever to have the result of the competitions, I think it is best to verify novice account that gets a medal (ask them for account verification and a code check), in a future competition give the possibility for the Kaggle community(preferably expert Kaggler) to report a user for a code check, for example, they get more than 20 reports or something like that the reported team will need to perform a code and identity check. Just trying to suggest some ideas
", "rawMarkdown": "I don't think it is possible for Kaggle team to check every single entry it will take forever to have the result of the competitions, I think it is best to verify novice account that gets a medal (ask them for account verification and a code check), in a future competition give the possibility for the Kaggle community(preferably expert Kaggler) to report a user for a code check, for example, they get more than 20 reports or something like that the reported team will need to perform a code and identity check. Just trying to suggest some ideas ", "votes": 3 }, { "id": 973940, "postDate": "2020-08-17T16:10:10.300Z", "content": "Great work @group16 . This kind of cheating will destroy Kaggle if let continue. It also seems like more screening is needed before users can participate in competitions that award medals and prize money.
", "rawMarkdown": "Great work @group16 . This kind of cheating will destroy Kaggle if let continue. It also seems like more screening is needed before users can participate in competitions that award medals and prize money. ", "votes": 3 }, { "id": 976771, "postDate": "2020-08-19T04:48:47.133Z", "content": "@group16, this is amazing. Well done.
\nI think Kaggle needs a team of private investigators to tackle this fairly new crime and you should be hired right away.
", "rawMarkdown": "@group16, this is amazing. Well done. \n\nI think Kaggle needs a team of private investigators to tackle this fairly new crime and you should be hired right away.", "votes": 4, "replies": [ { "id": 976817, "postDate": "2020-08-19T05:39:13.633Z", "content": "Thank you @sheriytm
\nDear @Kaggle, in case you have such an offer, I will gladly accept.
", "rawMarkdown": "Thank you @sheriytm \n\nDear @Kaggle, in case you have such an offer, I will gladly accept.", "votes": 6 }, { "id": 976825, "postDate": "2020-08-19T05:47:39.073Z", "content": "@group16,
\nPeut donner une référence si je dois.
My beginner's French at work.
", "rawMarkdown": "@group16, \nPeut donner une référence si je dois.\n\nMy beginner's French at work.", "votes": 1 }, { "id": 976865, "postDate": "2020-08-19T06:30:22.180Z", "content": "Haha merci beaucoup! En fait, je suis neerlandais, donc je pense que votre francais est meilleur de mien 🙈
", "rawMarkdown": "Haha merci beaucoup! En fait, je suis neerlandais, donc je pense que votre francais est meilleur de mien 🙈" }, { "id": 976919, "postDate": "2020-08-19T07:12:53.413Z", "content": "@sheriytm C'est en Wallonie qu'on parle français ^^
", "rawMarkdown": "@sheriytm C'est en [Wallonie](https://en.wikipedia.org/wiki/Wallonia) qu'on parle français ^^", "votes": 2 }, { "id": 976929, "postDate": "2020-08-19T07:21:03.923Z", "content": "Oui, c'est vrai! En Wallonie ils parlent francais. En Flandres, on parle neerlandais, francais et anglais. Francais est obligatoire dans les écoles flamands. Neerlandais n'est pas obligatoire dans les écoles francaises. Bienvenue en Belgique!
", "rawMarkdown": "Oui, c'est vrai! En Wallonie ils parlent francais. En Flandres, on parle neerlandais, francais et anglais. Francais est obligatoire dans les écoles flamands. Neerlandais **n'est pas** obligatoire dans les écoles francaises. Bienvenue en Belgique!", "votes": 3 }, { "id": 977338, "postDate": "2020-08-19T12:30:04.957Z", "content": "Thanks @serigne , I did not know Wallonia until you mentioned it. I just read it's history on Wikipedia. Amazing!
\n@group16,
\nJ'e doute que mon français soit meilleur que le vôtre. J'ai appris le français avec Duolingo.
Born surrounded by French speaking countries in West Africa, we were taught French in High school for the first 3yrs. I dropped it when we chose our elective subjects in the 4th year. I never practiced after that until I discovered the Duolingo app in 2017.
", "rawMarkdown": "Thanks @serigne , I did not know Wallonia until you mentioned it. I just read it's history on Wikipedia. Amazing!\n\n@group16, \nJ'e doute que mon français soit meilleur que le vôtre. J'ai appris le français avec [Duolingo](https://www.duolingo.com/).\n\nBorn surrounded by French speaking countries in West Africa, we were taught French in High school for the first 3yrs. I dropped it when we chose our elective subjects in the 4th year. I never practiced after that until I discovered the Duolingo app in 2017.", "votes": 1 }, { "id": 977395, "postDate": "2020-08-19T13:01:29.160Z", "content": "Yes, it can get quite complicated here in Belgium. There's Wallonia, Flanders and even a small German-speaking part. Then we have our capital Brussels as well which is its own region. Each of these have their own parliament. So that's 4 regional parliaments. We also have a federal parliament for the country. Very complicated in other words which is why we are breaking world records.
\nI've had french for 8 years on school and I could write (not speak) it pretty well. But that's more than 8 years ago by now so i've forgotten a lot…
", "rawMarkdown": "Yes, it can get quite complicated here in Belgium. There's Wallonia, Flanders and even a small German-speaking part. Then we have our capital Brussels as well which is its own region. Each of these have their own parliament. So that's 4 regional parliaments. We also have a federal parliament for the country. Very complicated in other words which is why we are breaking world records.\n\nI've had french for 8 years on school and I could write (not speak) it pretty well. But that's more than 8 years ago by now so i've forgotten a lot...", "votes": 2 }, { "id": 977459, "postDate": "2020-08-19T13:42:46.617Z", "content": "That is a lot complicated but learning many languages before the age of six connects so many neurons in a child's brain that proves super useful in life. I read a study on that years ago.
", "rawMarkdown": "That is a lot complicated but learning many languages before the age of six connects so many neurons in a child's brain that proves super useful in life. I read a study on that years ago. " }, { "id": 981257, "postDate": "2020-08-22T10:26:15.393Z", "content": "Le néerlandais est aussi obligatoire dans les écoles à Bruxelles avec au moins 4h de cours par semaines. Malheureusement pour moi car c'est une langue assez difficile.😆
\nNederlands is ook verplicht op school in Brussel met 4h per week. Helaas voor mij omdat het een heel moeilijke taal is.😆
", "rawMarkdown": "Le néerlandais est aussi obligatoire dans les écoles à Bruxelles avec au moins 4h de cours par semaines. Malheureusement pour moi car c'est une langue assez difficile.😆\n\nNederlands is ook verplicht op school in Brussel met 4h per week. Helaas voor mij omdat het een heel moeilijke taal is.😆", "votes": 1 }, { "id": 981276, "postDate": "2020-08-22T10:40:07.880Z", "content": "\n\nEn fait, je suis neerlandais
\n
Neerlandais ou Belge?
", "rawMarkdown": "> En fait, je suis neerlandais\n\nNeerlandais ou Belge?" }, { "id": 981281, "postDate": "2020-08-22T10:44:33.677Z", "rawMarkdown": "", "isDeleted": true }, { "id": 981290, "postDate": "2020-08-22T10:50:23.337Z", "content": "@cpmpml je suis Belge, je parle néerlandais/flamand :)
\n(Accidently deleted previous reply)
", "rawMarkdown": "@cpmpml je suis Belge, je parle néerlandais/flamand :)\n\n(Accidently deleted previous reply)", "votes": 1 }, { "id": 981309, "postDate": "2020-08-22T11:09:46.100Z", "content": "Je demande car j'ai un ami Neerlandais, et il me dit que les Flamands sont tres differents d'eux, meme si la langue est la meme….
", "rawMarkdown": "Je demande car j'ai un ami Neerlandais, et il me dit que les Flamands sont tres differents d'eux, meme si la langue est la meme...." }, { "id": 982006, "postDate": "2020-08-23T00:13:26.323Z", "content": "@cpmpml, @group16, - same language different people?
\nI can relate. I could speak 3 languages by the age of 5 - my language (1st), English (2nd), 3rd is the language most people around the area spoke. People often forget or are mistaken to think we all belong to the 3rd language.
", "rawMarkdown": "@cpmpml, @group16, - same language different people?\n\nI can relate. I could speak 3 languages by the age of 5 - my language (1st), English (2nd), 3rd is the language most people around the area spoke. People often forget or are mistaken to think we all belong to the 3rd language." } ] }, { "id": 975483, "postDate": "2020-08-18T10:10:38.483Z", "content": "People taking shortcuts will never learn.
", "rawMarkdown": "People taking shortcuts will never learn. ", "votes": 4, "replies": [ { "id": 975518, "postDate": "2020-08-18T10:32:44.733Z", "content": "Well, maybe they will learn that taking shortcuts isn't good. ;)
", "rawMarkdown": "Well, maybe they will **learn** that taking **shortcuts** isn't good. ;)", "votes": 2 }, { "id": 975614, "postDate": "2020-08-18T11:35:20.263Z", "rawMarkdown": "", "votes": 1, "isDeleted": true }, { "id": 975638, "postDate": "2020-08-18T11:48:17.890Z", "content": "Done + shortcuts as well!
", "rawMarkdown": "Done + shortcuts as well! " }, { "id": 975654, "postDate": "2020-08-18T12:03:13.133Z", "rawMarkdown": "", "votes": 2, "isDeleted": true }, { "id": 976199, "postDate": "2020-08-18T17:51:22.397Z", "content": "Awesome! Next step: Master. 😎
", "rawMarkdown": "Awesome! Next step: Master. 😎" }, { "id": 976678, "postDate": "2020-08-19T03:12:58.480Z", "rawMarkdown": "", "votes": 1, "isDeleted": true } ] }, { "id": 974014, "postDate": "2020-08-17T17:00:06.280Z", "content": "I think Kaggle needs to not allow novice account to join competitions until they become a Contributor. just to make it harder to cheat
", "rawMarkdown": "I think Kaggle needs to not allow novice account to join competitions until they become a Contributor. just to make it harder to cheat ", "votes": 4 }, { "id": 973941, "postDate": "2020-08-17T16:10:16.683Z", "content": "As someone who spent a ridiculous amount of time to barely scrape through to get my sole silver medal and competition expert status, this kind of behavior makes me incredibly sad. Thank you @group16 for bringing this to light and exposing these cheaters.
", "rawMarkdown": "As someone who spent a ridiculous amount of time to barely scrape through to get my sole silver medal and competition expert status, this kind of behavior makes me incredibly sad. Thank you @group16 for bringing this to light and exposing these cheaters.", "votes": 4 }, { "id": 973675, "postDate": "2020-08-17T12:58:20.060Z", "content": "Its scary to be precise. I hope these accounts get blocked.. @group16 - Hopefully, the particular person in your team has not been involved in private sharing, else your complete 3 months effort would be futile.. Appreciate your efforts in bringing this up.
", "rawMarkdown": "Its scary to be precise. I hope these accounts get blocked.. @group16 - Hopefully, the particular person in your team has not been involved in private sharing, else your complete 3 months effort would be futile.. Appreciate your efforts in bringing this up. ", "votes": 4 }, { "id": 973639, "postDate": "2020-08-17T12:20:24.833Z", "content": "@daishu wrote this previous year.
\n\n\nThere are some people who get solutions by joining different teams, and then improve their scores by private sharing. In China, we call this behavior \"PY transaction\", a typical group is \"宝典帮\". Now they are trying to pollute kaggle's competition environment, and they are sneaking into these teams:
\n
Those people might be joining multiple teams and collecting information from all of them just like before. Do you think your missing teammate can be in a circle like this? @group16
", "rawMarkdown": "@daishu wrote this previous year.\n\n> There are some people who get solutions by joining different teams, and then improve their scores by private sharing. In China, we call this behavior \"PY transaction\", a typical group is \"宝典帮\". Now they are trying to pollute kaggle's competition environment, and they are sneaking into these teams:\n\nThose people might be joining multiple teams and collecting information from all of them just like before. Do you think your missing teammate can be in a circle like this? @group16", "votes": 4, "replies": [ { "id": 973642, "postDate": "2020-08-17T12:22:59.713Z", "content": "I'm not sure but I am definitely not ruling that possibility out.
", "rawMarkdown": "I'm not sure but I am definitely not ruling that possibility out." } ] }, { "id": 973478, "postDate": "2020-08-17T10:20:20.677Z", "content": "yeah , so many novice accounts in gold medal range makes thing fishy
", "rawMarkdown": "yeah , so many novice accounts in gold medal range makes thing fishy", "votes": 4 }, { "id": 975931, "postDate": "2020-08-18T14:37:25.783Z", "content": "Makes me wonder what Data Science interviews in China are like.
", "rawMarkdown": "Makes me wonder what Data Science interviews in China are like.", "votes": 2, "replies": [ { "id": 976353, "postDate": "2020-08-18T19:59:41.103Z", "content": "@jackvial I wouldn't know, but I would imagine they are extremely difficult. Why? Because you have so many people who want those good jobs. You have an educated middle class growing at an astounding rate. The Chinese people (my experience with them while in higher ed) are extremely disciplined, hard working and competitive. So if some good DS job opens up, I bet its a shark tank.
", "rawMarkdown": "@jackvial I wouldn't know, but I would imagine they are extremely difficult. Why? Because you have so many people who want those good jobs. You have an educated middle class growing at an astounding rate. The Chinese people (my experience with them while in higher ed) are extremely disciplined, hard working and competitive. So if some good DS job opens up, I bet its a shark tank.", "votes": 11 }, { "id": 976586, "postDate": "2020-08-19T01:01:12.183Z", "content": "I have no doubt they are extremely difficult and there is a lot of competition. What I don't understand is how a market of buying and selling Kaggle medals is viable. Someone isn't going to get hired if they don't know anything about Data Science and just bought the medals, so are the medals only to make them more visible in the job market?
\nClearly there a lot of incredibly talented Data Scientists and Kagglers from China, I didn't mean any disrespect to them.
", "rawMarkdown": "I have no doubt they are extremely difficult and there is a lot of competition. What I don't understand is how a market of buying and selling Kaggle medals is viable. Someone isn't going to get hired if they don't know anything about Data Science and just bought the medals, so are the medals only to make them more visible in the job market?\n\nClearly there a lot of incredibly talented Data Scientists and Kagglers from China, I didn't mean any disrespect to them. ", "votes": 1 } ] }, { "id": 974060, "postDate": "2020-08-17T17:42:02.167Z", "content": "@divyansh22 I feel sorry for you. After so many years of education, how do you react to things? Blame the whole country? Please don't embarrass yourself.
", "rawMarkdown": "@divyansh22 I feel sorry for you. After so many years of education, how do you react to things? Blame the whole country? Please don't embarrass yourself.", "votes": 2, "replies": [ { "id": 974735, "postDate": "2020-08-18T03:15:53.203Z", "rawMarkdown": "", "votes": 1, "isDeleted": true } ] }, { "id": 973937, "postDate": "2020-08-17T16:08:00.047Z", "content": "gotta catch' em all!
", "rawMarkdown": "gotta catch' em all!", "votes": 2 }, { "id": 973932, "postDate": "2020-08-17T16:02:13.940Z", "content": "Thanks for you and your anonymous source's work. You've done the right thing. What happens in this competiton's public LB is CRAZING. Some people are getting more and more fearless that they now do not know what hiding means. A short and yet meanless way for money and fame drove them crazy.
\nI wish kaggle can take swift action in this competition along with long-term policies to stop things from happening again. I also understand that concrete evidence is extremely hard to find in this kind of situation and every policy relating to the whole community is hard to make. So thanks for kaggle team's great work behind the scene.
\nThe job needs to be done. Decaying and destroying is much more faster than growing and building. A snapshot of the current public LB can be a tragedy across the Internet. What kaggle does next matters in many ways: for kaggle itself; for organizers; for this wonderful community; for all other honest and devoted competitors; and for other communities in the spirit of fair sharing. The task could be heavy.
\nBest wish for kaggle and kagglers!
", "rawMarkdown": "Thanks for you and your anonymous source's work. You've done the right thing. What happens in this competiton's public LB is CRAZING. Some people are getting more and more fearless that they now do not know what hiding means. A short and yet meanless way for money and fame drove them crazy.\n\nI wish kaggle can take swift action in this competition along with long-term policies to stop things from happening again. I also understand that concrete evidence is extremely hard to find in this kind of situation and every policy relating to the whole community is hard to make. So thanks for kaggle team's great work behind the scene.\n\nThe job needs to be done. Decaying and destroying is much more faster than growing and building. A snapshot of the current public LB can be a tragedy across the Internet. **What kaggle does next matters in many ways: for kaggle itself; for organizers; for this wonderful community; for all other honest and devoted competitors; and for other communities in the spirit of fair sharing.** The task could be heavy.\n\nBest wish for kaggle and kagglers!", "votes": 3 }, { "id": 973545, "postDate": "2020-08-17T11:02:05.230Z", "content": "AFAIK the angtk.ai I knew was originally a data science community where people share general topics in data science (kinda like ods.ai) and help each other learn a few year back when I was still in high school. But then I got caught up in school work and didn't check their forum (and kaggle as well) for quite a while. Seems like they've evolved into this shady organization that profits (?) off of private sharing. What a pity. Or there may be two angtk.ai's and I was checking out the wrong one.
", "rawMarkdown": "AFAIK the angtk.ai I knew was originally a data science community where people share general topics in data science (kinda like ods.ai) and help each other learn a few year back when I was still in high school. But then I got caught up in school work and didn't check their forum (and kaggle as well) for quite a while. Seems like they've evolved into this shady organization that profits (?) off of private sharing. What a pity. Or there may be two angtk.ai's and I was checking out the wrong one.", "votes": 3, "replies": [ { "id": 973571, "postDate": "2020-08-17T11:26:10.833Z", "rawMarkdown": "", "votes": -3, "isDeleted": true } ] }, { "id": 974056, "postDate": "2020-08-17T17:40:19.807Z", "content": "I'm really really really sorry for my comment made on this post!!! I didn't hear what I was saying!! I really want to remove that comment but I am kinda unable to delete the discussion. There's a glitch with the discussions… I never intended to hurt anyone's feelings! I know I have hurt a lot of Kagglers feelings!! I'm really really sorry once again for my actions! No hard feelings on any specific community… I couldn't agree more that I sound really racist and I wish I could undo that! @kaggleteam, I really want to get things sorted, please look into the matter.🙏 Once again a deep regret and sorry for my actions!! I hope people forgive me, especially the ones I've offended!!🙏
", "rawMarkdown": "I'm really really really sorry for my comment made on this post!!! I didn't hear what I was saying!! I really want to remove that comment but I am kinda unable to delete the discussion. There's a glitch with the discussions... I never intended to hurt anyone's feelings! I know I have hurt a lot of Kagglers feelings!! I'm really really sorry once again for my actions! No hard feelings on any specific community... I couldn't agree more that I sound really racist and I wish I could undo that! @kaggleteam, I really want to get things sorted, please look into the matter.🙏 Once again a deep regret and sorry for my actions!! I hope people forgive me, especially the ones I've offended!!🙏", "votes": -1, "replies": [ { "id": 974126, "postDate": "2020-08-17T18:22:42.037Z", "content": "Glitch??? Yeah that \"glitch\" made you post hateful speech huh? And you never \"intended\" to hurt anyone? You know the worst and dangerous racist behaviors are being racist without awareness.
\nAll your bs here are just excuses, not an apology.
", "rawMarkdown": "Glitch??? Yeah that \"glitch\" made you post hateful speech huh? And you never \"intended\" to hurt anyone? You know the worst and dangerous racist behaviors are being racist without awareness. \n\nAll your bs here are just excuses, not an apology.", "votes": -3 }, { "id": 974170, "postDate": "2020-08-17T19:07:10.173Z", "content": "Hey @jihangz calm down, he meant to say there was a glitch in kaggle due to which he was unable to delete his hateful comment after self realization. I know he shouldn't have targeted a community but now he is accepting his mistake and apologized too. Maintain peace. ✌️
", "rawMarkdown": "Hey @jihangz calm down, he meant to say there was a glitch in kaggle due to which he was unable to delete his hateful comment after self realization. I know he shouldn't have targeted a community but now he is accepting his mistake and apologized too. Maintain peace. ✌️", "votes": 10 }, { "id": 974733, "postDate": "2020-08-18T03:13:11.847Z", "content": "Hi @jihangz, @nischaydnk is right about the glitch part. I was talking about Kaggle discussions glitch. Anyway, I'm really really sorry for the post I made earlier! I even tagged myself with Kaggle team to get my comment removed… Doesn't that mean I'm really sorry? It happened to me for the first time ever since I joined Kaggle, everybody who has been taking part with me in the competitions know how much I respect people…. You can ask @frtgnn, @muhakabartay, @yassinealouini, @ishivinal, @ruchi798, @dataraj. A lot of people know me outside Kaggle, you can ask them personally if you want… Once again I'm really sorry for my actions and i assure everyone that this won't happen anytime again in the future…🙏
", "rawMarkdown": "Hi @jihangz, @nischaydnk is right about the glitch part. I was talking about Kaggle discussions glitch. Anyway, I'm really really sorry for the post I made earlier! I even tagged myself with Kaggle team to get my comment removed... Doesn't that mean I'm really sorry? It happened to me for the first time ever since I joined Kaggle, everybody who has been taking part with me in the competitions know how much I respect people.... You can ask @frtgnn, @muhakabartay, @yassinealouini, @ishivinal, @ruchi798, @dataraj. A lot of people know me outside Kaggle, you can ask them personally if you want... Once again I'm really sorry for my actions and i assure everyone that this won't happen anytime again in the future...🙏", "votes": 2 } ] }, { "id": 980592, "postDate": "2020-08-21T17:55:45.083Z", "content": "I'm new to Kaggle and I'm glad I came across this post. It's really interesting that someone would buy a rank. It's kind of like someone buying a college degree without ever attending. When it comes down to it, if someone gets a job partially based on a Kaggle user rating, they may not be able to keep it due to lack of ability.
", "rawMarkdown": "I'm new to Kaggle and I'm glad I came across this post. It's really interesting that someone would buy a rank. It's kind of like someone buying a college degree without ever attending. When it comes down to it, if someone gets a job partially based on a Kaggle user rating, they may not be able to keep it due to lack of ability.", "votes": 1 }, { "id": 975542, "postDate": "2020-08-18T10:49:25.173Z", "content": "I always thought the purpose of this platform is to improve one's coding skill and try to make our best on difficult competitions… apparently not, i would never imagine such an industry could be possible…
", "rawMarkdown": "I always thought the purpose of this platform is to improve one's coding skill and try to make our best on difficult competitions... apparently not, i would never imagine such an industry could be possible...", "votes": 1, "replies": [ { "id": 975579, "postDate": "2020-08-18T11:13:14.597Z", "content": "Well it is still the case. Let's say that some are playing a different and worse game. 😉
", "rawMarkdown": "Well it is still the case. Let's say that some are playing a different and worse game. 😉 " }, { "id": 976344, "postDate": "2020-08-18T19:54:15.557Z", "content": "@bryanb well I am surprised that you are surprised. Humans exploit near every damn thing possible. Many of them are extremely desperate and will do desperate things. If they can find a way to not pay, not work, steal power, free food, cut corners………. do not underestimate the lazy, entitled human. Some of them are quite clever as well, and so the cat and mouse game ensues.
", "rawMarkdown": "@bryanb well I am surprised that you are surprised. Humans exploit near every damn thing possible. Many of them are extremely desperate and will do desperate things. If they can find a way to not pay, not work, steal power, free food, cut corners.......... do not underestimate the lazy, entitled human. Some of them are quite clever as well, and so the cat and mouse game ensues.", "votes": 1 } ] }, { "id": 974142, "postDate": "2020-08-17T18:38:51.080Z", "content": "My god, the evidence just keeps mounting. Now we have a confession of a Kaggle master in the discussions. Basically anyone associated with angtk.com seems part of this ring. Either you are being paid, as a Master/GM, to take on newbies into your team, who can contribute nothing, so they can get medals……….you can claim ignorance……and that you were just doing it for advertising. There also appears to be code sharing/selling which is distributed through some private method, Kaggle accounts are created with no bio, no activity, they do not participate in discussions, comments, or anything. These accounts submit top scores…..they could be used to lower others, or to probe or to try to go for medals. Some of the accounts have never been used yet. All associated with angtk.com.
", "rawMarkdown": "My god, the evidence just keeps mounting. Now we have a confession of a Kaggle master in the discussions. Basically anyone associated with angtk.com seems part of this ring. Either you are being paid, as a Master/GM, to take on newbies into your team, who can contribute nothing, so they can get medals..........you can claim ignorance......and that you were just doing it for advertising. There also appears to be code sharing/selling which is distributed through some private method, Kaggle accounts are created with no bio, no activity, they do not participate in discussions, comments, or anything. These accounts submit top scores.....they could be used to lower others, or to probe or to try to go for medals. Some of the accounts have never been used yet. All associated with angtk.com.", "votes": 1 }, { "id": 974044, "postDate": "2020-08-17T17:33:35.553Z", "content": "@group16 Its sad that there are people using kaggle to sell medals and not to learn. At least there are still people using kaggle to learn and not to cheat.
", "rawMarkdown": "@group16 Its sad that there are people using kaggle to sell medals and not to learn. At least there are still people using kaggle to learn and not to cheat.", "votes": 1 }, { "id": 974011, "postDate": "2020-08-17T16:59:32.657Z", "content": "@stanleyjzheng Kaggle discussions are broken. I hope this gets fixed.
", "rawMarkdown": "@stanleyjzheng Kaggle discussions are broken. I hope this gets fixed.", "votes": 1 }, { "id": 973890, "postDate": "2020-08-17T15:30:21.447Z", "content": "That's ridiculous! It even doesn't make any sense since people learn NOTHING from these kind of sources, just ruin the other's competition experience
", "rawMarkdown": "That's ridiculous! It even doesn't make any sense since people learn NOTHING from these kind of sources, just ruin the other's competition experience", "votes": 1 }, { "id": 973874, "postDate": "2020-08-17T15:17:32.690Z", "content": "this is pretty scary, thanks for sharing. Hope you can get a better teammate who wont ghost you
", "rawMarkdown": "this is pretty scary, thanks for sharing. Hope you can get a better teammate who wont ghost you", "votes": 1 }, { "id": 973843, "postDate": "2020-08-17T14:54:30.960Z", "content": "kaggle can implement something like an user account should be old enough before joining any competitions.
", "rawMarkdown": "kaggle can implement something like an user account should be old enough before joining any competitions.", "votes": 1 }, { "id": 973794, "postDate": "2020-08-17T14:16:50.397Z", "content": "If Kaggle workers don't ban them now they will do it over and over again!
", "rawMarkdown": "If Kaggle workers don't ban them now they will do it over and over again!", "votes": 1 }, { "id": 973681, "postDate": "2020-08-17T13:04:34.467Z", "content": "Kaggle has to take a decision regarding this issue bt tracing the Profiles, it doesn't make sense that with only few numbers of submissions less than 50 submissions to reach first five.
", "rawMarkdown": "Kaggle has to take a decision regarding this issue bt tracing the Profiles, it doesn't make sense that with only few numbers of submissions less than 50 submissions to reach first five.", "votes": 1 }, { "id": 973679, "postDate": "2020-08-17T13:03:17.817Z", "content": "So, if I understand correctly, they use these fake accounts for testing submissions. They choose other (usually low-score submits for final score) and they sell high-scores submissions to the third party.
\nYou raised a very important question, let's solve it together. Maybe Kaggle workers will be forced to rebuild the system of checking users and their submissions. To be honest, there is no system at all. It's time to build it.
\nThere are two ways how to fight with that:
\n1) Find out who is behind all this and punish them.
\n2) Find out cheating solutions and ban them.
\nI think the first one is much more effective. We should fight not with a cause but with a reason.
It also works the other way around.
\nSee if you have students who will pay for submissions… it gives the cheaters the advantage of improving their scores as they can use the information from the public lb of their students.
The students are actually farmers who rented a piece of land and must also share some of their crops with their landlords.
", "rawMarkdown": "It also works the other way around. \nSee if you have students who will pay for submissions... it gives the cheaters the advantage of improving their scores as they can use the information from the public lb of their students.\n\nThe students are actually farmers who rented a piece of land and must also share some of their crops with their landlords.", "votes": 1 }, { "id": 973598, "postDate": "2020-08-17T11:46:30.480Z", "content": "Thanks for raising awareness on this! I did not know such a phenomenon existed in Kaggle or rather I didn't pay much attention to it until now. I think the most important thing is to learn as a community, we should come up with ways of detecting such misbehavior and penalize these users
", "rawMarkdown": "Thanks for raising awareness on this! I did not know such a phenomenon existed in Kaggle or rather I didn't pay much attention to it until now. I think the most important thing is to learn as a community, we should come up with ways of detecting such misbehavior and penalize these users", "votes": 1 }, { "id": 973523, "postDate": "2020-08-17T10:48:25.127Z", "content": "I remember something similar happened in the IEEE-CIS Fraud detection competition (almost a year ago) as well.
", "rawMarkdown": "I remember something similar happened in the IEEE-CIS Fraud detection competition (almost a year ago) as well. ", "votes": 1, "replies": [ { "id": 975922, "postDate": "2020-08-18T14:24:37.427Z", "rawMarkdown": "", "isDeleted": true } ] }, { "id": 974789, "postDate": "2020-08-18T03:53:43.770Z", "content": "Listen straight from the horse's mouth
\n\n\nI've given some notebook and docs from past competitions as examples. And they pay me for this. At the same time, they asked me to add the names of some of the teams[ angtk.ai ]. pay for this. But I don't know what kind of ads they use these screenshots for. At the same time, advertisements from more than one source cannot be used as evidence.
\n
https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/175201
\n\n\nI don't know where these screenshots came from. I've shared with you some of the cases that have ended the competition angtk.ai They also asked me to add the team name[ angtk.ai ]And pay for it. I didn't do anything else. Which screenshots came from, or from PS, can't be confirmed.
\n
https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/175196
\n\n\nI sold them some code that had ended the game as a case study. They asked me to put it before the team name[ angtk.ai ]As an advertisement and pay for it! They used a screenshot of my team as an ad, but I didn't confirm the content carefully. I think it's really strange to say that I have private use with some screenshots of advertisements. In China, most advertisements exaggerate!
\n
https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/175205
", "rawMarkdown": "**Listen straight from the horse's mouth**\n\n> I've given some notebook and docs from past competitions as examples. And they pay me for this. At the same time, they asked me to add the names of some of the teams[ angtk.ai ]. pay for this. But I don't know what kind of ads they use these screenshots for. At the same time, advertisements from more than one source cannot be used as evidence.\n\nhttps://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/175201\n\n> I don't know where these screenshots came from. I've shared with you some of the cases that have ended the competition angtk.ai They also asked me to add the team name[ angtk.ai ]And pay for it. I didn't do anything else. Which screenshots came from, or from PS, can't be confirmed.\n\nhttps://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/175196\n\n> I sold them some code that had ended the game as a case study. They asked me to put it before the team name[ angtk.ai ]As an advertisement and pay for it! They used a screenshot of my team as an ad, but I didn't confirm the content carefully. I think it's really strange to say that I have private use with some screenshots of advertisements. In China, most advertisements exaggerate!\n\nhttps://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/175205\n\n", "votes": 2 }, { "id": 974514, "postDate": "2020-08-18T01:04:30.347Z", "content": "Based on the results, Now we can say that Public LB is totally screwed up but \"Hard work will never fail\". I like that the Kaggle team has worked on and yah still working on removing violaters.
", "rawMarkdown": "Based on the results, Now we can say that Public LB is totally screwed up but **\"Hard work will never fail\".** I like that the Kaggle team has worked on and yah still working on removing violaters.", "votes": 2, "replies": [ { "id": 974527, "postDate": "2020-08-18T01:12:58.160Z", "rawMarkdown": "", "isDeleted": true }, { "id": 974530, "postDate": "2020-08-18T01:15:07.360Z", "content": "KARMA never loses an address.
", "rawMarkdown": "**KARMA** never loses an address.", "votes": 2 } ] }, { "id": 974081, "postDate": "2020-08-17T18:00:41.037Z", "content": "@group16 thanks for the breakdown and here and in https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/174635#973667 . This is no small problem. In this competition alone it seems there may be dozens of people farming medals, many at the top of the LB.
", "rawMarkdown": "@group16 thanks for the breakdown and here and in https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/174635#973667 . This is no small problem. In this competition alone it seems there may be dozens of people farming medals, many at the top of the LB.", "votes": 2 }, { "id": 973989, "postDate": "2020-08-17T16:46:05.763Z", "content": "Oh my god, I double checked if you were Chinese.😹
\nThis is really disgusting to buy a Kaggle medal! It made so many people's hard work end up with nothing.
Vopani\"s words are wise:
\n\n", "rawMarkdown": "Vopani\"s words are wise: \n\n> Anything that has monetary gains will have scammers " }, { "id": 974070, "postDate": "2020-08-17T17:52:17.743Z", "content": "Anything that has monetary gains will have scammers
\n
@yuanlin08 I have nothing to do with education, it just the perfect example for when u talk out of anger, he did apologize for it no need to be that hard on him.
", "rawMarkdown": "@yuanlin08 I have nothing to do with education, it just the perfect example for when u talk out of anger, he did apologize for it no need to be that hard on him.", "votes": 1, "replies": [ { "id": 974146, "postDate": "2020-08-17T18:42:43.217Z", "content": "He apologized after i posted, thanks.
", "rawMarkdown": "He apologized after i posted, thanks. ", "votes": 2 }, { "id": 974794, "postDate": "2020-08-18T03:56:27.630Z", "content": "Sometimes it just leaks what's rooted in your mindset..
", "rawMarkdown": "Sometimes it just leaks what's rooted in your mindset.." } ] }, { "id": 978980, "postDate": "2020-08-20T14:50:29.133Z", "content": "nice very!!!!!
", "rawMarkdown": "nice very!!!!!", "votes": -5 }, { "id": 979201, "postDate": "2020-08-20T17:29:07.370Z", "content": "nice……
", "rawMarkdown": "nice......", "votes": -4 }, { "id": 978911, "postDate": "2020-08-20T13:47:22.027Z", "content": "@group16 , Great !!!
", "rawMarkdown": "@group16 , Great !!!", "votes": -6 }, { "id": 987357, "postDate": "2020-08-27T07:29:40.653Z", "content": "I'm new to kaggle but buying ranks for fake knowledge depiction would never ever work in life.
", "rawMarkdown": "I'm new to kaggle but buying ranks for fake knowledge depiction would never ever work in life.", "replies": [ { "id": 987390, "postDate": "2020-08-27T08:12:04.360Z", "content": "It happens all the time in life. People trade money for power, sex for power…….it's not honorable, and certainly has no place here, but corruption infiltrates everything.
", "rawMarkdown": "It happens all the time in life. People trade money for power, sex for power.......it's not honorable, and certainly has no place here, but corruption infiltrates everything." } ] }, { "id": 986630, "postDate": "2020-08-26T16:33:05.913Z", "content": "I'm new to Kaggle and I'm glad I came across this post good job
", "rawMarkdown": "I'm new to Kaggle and I'm glad I came across this post good job " }, { "id": 985889, "postDate": "2020-08-26T04:55:43.120Z", "content": "Kaggle Please take some actions on these type of fraud practices.
", "rawMarkdown": "Kaggle Please take some actions on these type of fraud practices.\n" }, { "id": 981997, "postDate": "2020-08-22T23:55:29.303Z", "content": "maybe Kaggle admins can use anomaly detection tech based on user's notebook logs. Very few cases can get a high mark in first-run unless someone copied the code or model.
", "rawMarkdown": "maybe Kaggle admins can use anomaly detection tech based on user's notebook logs. Very few cases can get a high mark in first-run unless someone copied the code or model. " }, { "id": 979068, "postDate": "2020-08-20T15:40:46.723Z", "content": "I want to see write-ups of solution from teams @siim2020_stacking, @sunlight and @zhaigongzuo2
", "rawMarkdown": "I want to see write-ups of solution from teams @siim2020_stacking, @sunlight and @zhaigongzuo2" }, { "id": 975895, "postDate": "2020-08-18T14:09:57.067Z", "content": "It is so awful, kaggle maybe improve the competition enviornment.
", "rawMarkdown": "It is so awful, kaggle maybe improve the competition enviornment." }, { "id": 974367, "postDate": "2020-08-17T23:48:14.737Z", "content": "Thank you for heads up.
\nReally like the idea of metadata graphing too.
\nAfter comp deadline, before any results published or winners announced and remove all cheats.
I guess this website is also used to share content? https://www.angtk.com/
", "rawMarkdown": "I guess this website is also used to share content? https://www.angtk.com/", "replies": [ { "id": 974274, "postDate": "2020-08-17T21:02:17.507Z", "content": "probably. The sub domain edu.angtk.com was where these $1000 Kaggle \"classes\" were happening.
", "rawMarkdown": "probably. The sub domain edu.angtk.com was where these $1000 Kaggle \"classes\" were happening.", "votes": 1 }, { "id": 976932, "postDate": "2020-08-19T07:22:11.617Z", "content": "Hey, the website seems to be down now ?
", "rawMarkdown": "Hey, the website seems to be down now ?" } ] }, { "id": 974212, "postDate": "2020-08-17T19:48:20.817Z", "content": "Can I assume this type of fraud activity is one of the kind of reasons why genuine kaggle grandmasters or genuine competitors leaving this place to compete?
", "rawMarkdown": "Can I assume this type of fraud activity is one of the kind of reasons why genuine kaggle grandmasters or genuine competitors leaving this place to compete?" }, { "id": 974085, "postDate": "2020-08-17T18:02:11.370Z", "content": "I have a feeling that we will wait a lot more than usual for the final private lb and medals :)
", "rawMarkdown": "I have a feeling that we will wait a lot more than usual for the final private lb and medals :) " }, { "id": 974020, "postDate": "2020-08-17T17:04:29.267Z", "content": "Thanks for this information :) This is quite a revelation for someone who started a week ago. Hope the community finds a way to circumvent this.
", "rawMarkdown": "Thanks for this information :) This is quite a revelation for someone who started a week ago. Hope the community finds a way to circumvent this. " }, { "id": 973947, "postDate": "2020-08-17T16:14:03.370Z", "content": "Shame on them
", "rawMarkdown": "Shame on them" }, { "id": 973568, "postDate": "2020-08-17T11:23:11.843Z", "content": "You should work at kaggle's compliance department :)
\nOne might think that it would be easy for Google/Kaggle with its resources to find cheaters, especially if someone flag with a name where a solution was bought if it is as you say. And why signal the misstake in the name, all strange.
\nHave you reported the teams / profiles / notebooks to Kaggle? then they can certainly do their thorough investigation.
It is very difficult to catch all, specifically in csv only competitions. When I was teaching in university the only way to catch all students plagiarizing is to schedule personal rehearsals and let them describe the solution, then it is super easy to identify who did what and understands what he or she did.
", "rawMarkdown": "It is very difficult to catch all, specifically in csv only competitions. When I was teaching in university the only way to catch all students plagiarizing is to schedule personal rehearsals and let them describe the solution, then it is super easy to identify who did what and understands what he or she did.", "votes": 19, "replies": [ { "id": 973611, "postDate": "2020-08-17T11:57:33.840Z", "content": "May be all gold or even silver solutions need to be public . If not the full code ?
", "rawMarkdown": "May be all gold or even silver solutions need to be public . If not the full code ? ", "votes": 1 }, { "id": 973621, "postDate": "2020-08-17T12:04:36.083Z", "content": "Yes, only kernel competitions and everything needs to be open source would be at least a step into the right direction. However, training would not be part of that, so still an issue.
", "rawMarkdown": "Yes, only kernel competitions and everything needs to be open source would be at least a step into the right direction. However, training would not be part of that, so still an issue.", "votes": 5 }, { "id": 973625, "postDate": "2020-08-17T12:08:50.613Z", "content": "Agree. But I don't think the organizers have time to consider \"Not in the money\" solutions, because they're only interested in the top solution anyway.
", "rawMarkdown": "Agree. But I don't think the organizers have time to consider \"Not in the money\" solutions, because they're only interested in the top solution anyway.", "votes": 3 } ] }, { "id": 973581, "postDate": "2020-08-17T11:33:36.527Z", "content": "Wow do you actually spend time catching them? You'd make a responsible professor. @philippsinger
", "rawMarkdown": "Wow do you actually spend time catching them? You'd make a responsible professor. @philippsinger ", "votes": 1 }, { "id": 973585, "postDate": "2020-08-17T11:38:28.113Z", "content": "Yes, it was common in the university I was studying, but not common in others. Students were not expecting it… I find it important for them to learn that this is an absolute no-go
", "rawMarkdown": "Yes, it was common in the university I was studying, but not common in others. Students were not expecting it... I find it important for them to learn that this is an absolute no-go", "votes": 4 } ] }, { "id": 981181, "postDate": "2020-08-22T08:46:17Z", "rawMarkdown": "", "votes": -2, "isDeleted": true }, { "id": 979451, "postDate": "2020-08-20T21:09:26.007Z", "rawMarkdown": "", "isDeleted": true }, { "id": 976904, "postDate": "2020-08-19T07:04:32.097Z", "rawMarkdown": "", "votes": -1, "isDeleted": true, "replies": [ { "id": 976907, "postDate": "2020-08-19T07:07:07.460Z", "content": "I was already wondering how long it would take the bots to arrive here.
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F443651%2Fa8dbec437dd6e020e8f42940d8501f3f%2Fso-it-begins-5b8182.jpg?generation=1597820818288955&alt=media\")
Don't worry about it! Participate in discussions, do competitions, release notebooks. Novices are a great addition to the community; we all must start somewhere. Just don't get top 10 in a competition with no discussion history, no notebooks, and no activity; that's quite fishy.
\nAlso please consider putting in a bit of time to get contributor rank, it doesn't take long at all.
", "rawMarkdown": "Don't worry about it! Participate in discussions, do competitions, release notebooks. Novices are a great addition to the community; we all must start somewhere. Just don't get top 10 in a competition with no discussion history, no notebooks, and no activity; that's quite fishy.\n\nAlso please consider putting in a bit of time to get contributor rank, it doesn't take long at all. ", "votes": 1 } ] }, { "id": 974009, "postDate": "2020-08-17T16:57:42.250Z", "rawMarkdown": "", "isDeleted": true }, { "id": 973761, "postDate": "2020-08-17T13:55:48.067Z", "rawMarkdown": "", "votes": -5, "isDeleted": true, "replies": [ { "id": 973788, "postDate": "2020-08-17T14:13:42.577Z", "content": "I disagree you. Chinese competitors are not responsible for this small group of people.
", "rawMarkdown": "I disagree you. Chinese competitors are not responsible for this small group of people.", "votes": 11 } ] }, { "id": 973678, "postDate": "2020-08-17T13:02:53.253Z", "rawMarkdown": "", "isDeleted": true }, { "id": 973677, "postDate": "2020-08-17T13:02:22.380Z", "rawMarkdown": "", "isDeleted": true }, { "id": 973578, "postDate": "2020-08-17T11:29:58.783Z", "rawMarkdown": "", "votes": -5, "isDeleted": true, "replies": [ { "id": 973582, "postDate": "2020-08-17T11:34:36.163Z", "rawMarkdown": "", "votes": 2, "isDeleted": true }, { "id": 973583, "postDate": "2020-08-17T11:36:54.203Z", "rawMarkdown": "", "votes": -1, "isDeleted": true }, { "id": 973584, "postDate": "2020-08-17T11:36:55.500Z", "content": "I got them from an anonymous source. But yes, they could have been taken from a computer and (different) phones. That does not matter right? These screenshots have been collected over a long period of time, not all in a single moment.
\nRegarding the angtk organization, I will rephrase that to make sure I don't throw false accusations.
", "rawMarkdown": "I got them from an anonymous source. But yes, they could have been taken from a computer and (different) phones. That does not matter right? These screenshots have been collected over a long period of time, not all in a single moment.\n\nRegarding the angtk organization, I will rephrase that to make sure I don't throw false accusations. ", "votes": 2 }, { "id": 973586, "postDate": "2020-08-17T11:39:18.970Z", "rawMarkdown": "", "votes": -1, "isDeleted": true }, { "id": 973588, "postDate": "2020-08-17T11:40:08.020Z", "rawMarkdown": "", "votes": 1, "isDeleted": true }, { "id": 973591, "postDate": "2020-08-17T11:40:48.457Z", "content": "Not sure. Here's the link but I would need to translate to make sense of what's on that page :).
", "rawMarkdown": "Not sure. [Here's the link](https://shop319532120.world.taobao.com/?spm=a230r.7195193.1997079397.2.52a85a59xgPAS6) but I would need to translate to make sense of what's on that page :).", "votes": 1 }, { "id": 973594, "postDate": "2020-08-17T11:43:41.450Z", "rawMarkdown": "", "votes": -2, "isDeleted": true }, { "id": 973600, "postDate": "2020-08-17T11:48:14.690Z", "content": "I'm sorry but you can let others judge please
", "rawMarkdown": "I'm sorry but you can let others judge please", "votes": 1 }, { "id": 973602, "postDate": "2020-08-17T11:50:38.783Z", "content": "so many teams advanced 2 places, is that interesting?
", "rawMarkdown": "so many teams advanced 2 places, is that interesting?", "votes": 2 }, { "id": 973603, "postDate": "2020-08-17T11:51:05.917Z", "rawMarkdown": "", "isDeleted": true }, { "id": 973607, "postDate": "2020-08-17T11:53:00.717Z", "content": "I didn't say you are a cheater. But you defended for angtk seems quite strange when everyone agree that they have problems.
", "rawMarkdown": "I didn't say you are a cheater. But you defended for angtk seems quite strange when everyone agree that they have problems.", "votes": 3 }, { "id": 973610, "postDate": "2020-08-17T11:54:46.483Z", "rawMarkdown": "", "votes": 1, "isDeleted": true } ] }, { "id": 976704, "postDate": "2020-08-19T03:48:05.437Z", "content": "Cool. Thank you.
", "rawMarkdown": "Cool. Thank you.", "votes": -4 } ], "comments": [ { "id": 973631, "author_name": "Serigne ", "author_url": "", "post_date": "2020-08-17T12:14:17.677000", "content": "Almost all the golden area is overtaken by novice cheaters accounts.
\nIf they're still there on private LB, I will likely never compete again on kaggle.
", "votes": 44, "replies": [ { "id": 973662, "author_name": "Tim Yee", "author_url": "", "post_date": "2020-08-17T12:46:41.907000", "content": "the [angtk.ai]poxiao team was not removed. they simply dropped 11 spots to finish 15 on private LB and renamed to their team to oxiao.
", "votes": 3, "replies": [] }, { "id": 973669, "author_name": "Vopani", "author_url": "", "post_date": "2020-08-17T12:53:26.957000", "content": "Meant for the ongoing Melanoma competition.
", "votes": 1, "replies": [] }, { "id": 973741, "author_name": "Kha Vo", "author_url": "", "post_date": "2020-08-17T13:44:20.153000", "content": "That is not right. You cannot commit suicide just because you don’t like people around you. Stay with it and we’ll have a solution together, or at least there are people who support you.
", "votes": 11, "replies": [] }, { "id": 973789, "author_name": "Serigne ", "author_url": "", "post_date": "2020-08-17T14:14:32.640000", "content": "Your comparison does not seem very appropriate to me.
\nThey are apparently serial cheaters who have been doing it for a long time. Faced with such a body of evidences gathered and shared here, if kaggle doesn't sanction them, I don't see what would keep them from doing it over and over again, and even inciting others to do the same given there would be no sanction. And the competitions will lose all their interest.
", "votes": 5, "replies": [] }, { "id": 973815, "author_name": "Kha Vo", "author_url": "", "post_date": "2020-08-17T14:26:58.297000", "content": "Just be here, don't go away. You are a passionate data scientist so don't flee away from anything
", "votes": 3, "replies": [] } ] }, { "id": 973698, "author_name": "Andrey Lukyanenko", "author_url": "", "post_date": "2020-08-17T13:14:15.740000", "content": "I hope Kaggle will ban all of those who buy or sell medals.
", "votes": 30, "replies": [] }, { "id": 974218, "author_name": "CPMP", "author_url": "", "post_date": "2020-08-17T19:54:20.337000", "content": "You have a future in Intelligence industry ;)
", "votes": 26, "replies": [ { "id": 974219, "author_name": "Gilles Vandewiele", "author_url": "", "post_date": "2020-08-17T19:56:13.757000", "content": "No word from the Belgian government yet unfortunately… They might be too busy forming a government.
", "votes": 30, "replies": [] }, { "id": 974249, "author_name": "Yassine Alouini", "author_url": "", "post_date": "2020-08-17T20:34:10.970000", "content": "Are they still choosing the government? Is that another 2010-11 crisis? :D
\nhttps://en.wikipedia.org/wiki/2010%E2%80%9311_Belgian_government_formation
We broke our own world record :)
\n\n", "votes": 2, "replies": [] }, { "id": 974267, "author_name": "Yassine Alouini", "author_url": "", "post_date": "2020-08-17T20:56:19.333000", "content": "That's both funny and sad. Vive la Belgique. :)
", "votes": 2, "replies": [] }, { "id": 974283, "author_name": "Serigne ", "author_url": "", "post_date": "2020-08-17T21:11:06.993000", "content": "At least you have fries ;)
", "votes": 2, "replies": [] }, { "id": 974286, "author_name": "Yassine Alouini", "author_url": "", "post_date": "2020-08-17T21:13:03.043000", "content": "And gaufre. 😍
", "votes": 2, "replies": [] }, { "id": 974299, "author_name": "Serigne ", "author_url": "", "post_date": "2020-08-17T21:20:07.767000", "content": "Hopefully Americans don't call them \"french gaufre\" (like french fries ^^)
\nBut I have not much hope for that , given for Trump : \"Belgium is a beautiful city\" :)
", "votes": 1, "replies": [] }, { "id": 974307, "author_name": "Yassine Alouini", "author_url": "", "post_date": "2020-08-17T21:28:12.283000", "content": "Well, given its size… 😏
", "votes": 1, "replies": [] }, { "id": 975393, "author_name": "Gilles Vandewiele", "author_url": "", "post_date": "2020-08-18T09:18:53.087000", "content": "Petit mais parfait! Don't be jealous of our national treasures guys: beers, waffles, chocolate, fries, …
\nYou french people still have your fermented grape juice and moldy milk!
", "votes": 4, "replies": [] }, { "id": 975411, "author_name": "Gajendra Saraswat", "author_url": "", "post_date": "2020-08-18T09:32:24.830000", "content": "I just searched \"10 things Belgium is famous for\", the first thing was the highest number of Chocolate sales of course, but the second one is this: 😄😄
\n
I didn't know it was that big of a achievement lol!
", "votes": 2, "replies": [] }, { "id": 975532, "author_name": "Gilles Vandewiele", "author_url": "", "post_date": "2020-08-18T10:42:34.523000", "content": "It's one of our greatest national prides. A country as small as ours doesn't get to seize a world record every day… Do you have the article? I'm intrigued by point 3 as well (party freaks, which are references to Tomorrowland I guess)
\nEDIT: Also: \"Everything ran smoothly during this era\" 🙈🙈🙈
", "votes": 2, "replies": [ { "id": 975573, "author_name": "Nirjhar Roy", "author_url": "", "post_date": "2020-08-18T11:09:34.677000", "content": "@group16 Everything ran smoothly during that era . Ofcourse ! Point 1 is chocolate and point 3 is party freak ! Nothing can go wrong there :) People are happy ..
", "votes": 1, "replies": [] } ] }, { "id": 975576, "author_name": "Gajendra Saraswat", "author_url": "", "post_date": "2020-08-18T11:12:11.413000", "content": "Ohh God, Tomorrow land is from Belgium? I didn't knew that either, what am I doing in my life. :/
\nCool, here is your Pride. 😄
\nBtw, in our state, our govt also moved to a hotel for some picnic stuff for a month and we were without any government too, although that's a different thing that we don't want them either! 🤕🤕
\nP.S.: I read the article again, Belgium really is amazing!
", "votes": 1, "replies": [] }, { "id": 975646, "author_name": "Gilles Vandewiele", "author_url": "", "post_date": "2020-08-18T11:57:24.957000", "content": "That was a fun article, many thanks. I've drank some beers in 'T Brugs Beertje that can be seen in one of the pictures (which is located in my home-city).
", "votes": 1, "replies": [] } ] }, { "id": 973849, "author_name": "Heads or Tails", "author_url": "", "post_date": "2020-08-17T15:00:56.650000", "content": "Thank you, Gilles and your anonymous source, for exposing those fraudulent schemes. These methods constitute such a blatant disrespect of the community; it is very disappointing to see. I hope that the Kaggle team will take swift and decisive actions against them. The hard work and effort of all honest competitors should not be undermined like this.
\nTo Kaggle's credit, we know that accounts of cheaters are regularly removed, and that a lot of work is being done behind the scenes to keep the community clean. In the past, even cheating \"Grandmasters\" have been banned. However, those higher-profile cases normally take quite a lot of time, during which the bad actors can get away with more cheating. While I can understand that Kaggle wants to err on the side of caution, time is really of the essence here. A lot of damage is being done if groups of people can get away with fraud for even 1 competition, since that invites copycat behaviour.
\nAs Kaggle continues to grow in popularity, anti-fraud actions need to be taken faster. Otherwise our community might gradually lose its hard-earned reputation.
\nI would also like to say, Gilles, that you and your honest teammates did nothing wrong during this competition. You did your due diligence before teaming up, and in a better situation this shady account would likely have never had the results to make it seem trustworthy. I hope that Kaggle will find a way to punish the cheaters without it affecting the honest competitors.
", "votes": 24, "replies": [] }, { "id": 974013, "author_name": "Changyi", "author_url": "", "post_date": "2020-08-17T16:59:59.570000", "content": "@divyansh22 if I see a cheater from your country, I would never say your country is full of cheaters.
\nPlease pay attention, we're talking about cheating not racism.
", "votes": 22, "replies": [ { "id": 974724, "author_name": "", "author_url": "", "post_date": "2020-08-18T03:05:19.887000", "content": "", "votes": 1, "replies": [] }, { "id": 974785, "author_name": "", "author_url": "", "post_date": "2020-08-18T03:52:05.507000", "content": "I'm not surprised that some hateful words come from our neighbor…
", "votes": 0, "replies": [] } ] }, { "id": 973497, "author_name": "beluga", "author_url": "", "post_date": "2020-08-17T10:30:50.353000", "content": "Fortunately I did not delete my WeChat app after Kaggle Days in Beijing. Let's go for shopping!
\nBtw I also have 11 silver medals and a few of them I don't need anymore if anyone interested :D
", "votes": 18, "replies": [ { "id": 973501, "author_name": "Gilles Vandewiele", "author_url": "", "post_date": "2020-08-17T10:31:42.843000", "content": "What's your offer? :D
", "votes": 2, "replies": [] }, { "id": 973503, "author_name": "Gajendra Saraswat", "author_url": "", "post_date": "2020-08-17T10:34:51.863000", "content": "You wanna put them on sale? 😂😂
", "votes": 0, "replies": [] }, { "id": 973528, "author_name": "earhian", "author_url": "", "post_date": "2020-08-17T10:50:25.013000", "content": "", "votes": 2, "replies": [] }, { "id": 974185, "author_name": "", "author_url": "", "post_date": "2020-08-17T19:24:52.690000", "content": "", "votes": 0, "replies": [] }, { "id": 974753, "author_name": "", "author_url": "", "post_date": "2020-08-18T03:29:04.543000", "content": "Any \"Kaggler discount coupon\" for that? 😅
", "votes": 0, "replies": [ { "id": 974791, "author_name": "Signal", "author_url": "", "post_date": "2020-08-18T03:54:57.090000", "content": "@stevenyin I do have one discount code, but please don't share it, I had to pay for it. Its code TEAMYIKOUOUBAO and you can use it at any of angtk.com's various Kaggle markets.
", "votes": 2, "replies": [] } ] }, { "id": 974818, "author_name": "", "author_url": "", "post_date": "2020-08-18T04:11:24.720000", "content": "Just tried it and only found already expired. I must be too slow to use it..☹️
", "votes": 0, "replies": [] } ] }, { "id": 973790, "author_name": "عثمان", "author_url": "", "post_date": "2020-08-17T14:14:40.287000", "content": "For those who’ve worked in the fraud detection space, it’s so interesting because transaction time (or submission time in our case) is such an interesting feature. Intuitively, you’d think they’d stagger their submissions to make the fraud seem more organic, but that goes against another driving factor, which is the desire to make as much money off the scheme as quickly as possible before the lid gets blown off the operation.
\n🍿🍿🍿🍿🍿
", "votes": 15, "replies": [] }, { "id": 973736, "author_name": "Waylon Wu", "author_url": "", "post_date": "2020-08-17T13:38:12.217000", "content": "Nice work, @group16 . Kaggle needs you, the world needs you. :D
", "votes": 15, "replies": [] }, { "id": 974159, "author_name": "Dracarys", "author_url": "", "post_date": "2020-08-17T18:57:57.700000", "content": "On a lighter note, i think it's time we will soon see a kaggle competition about detecting fraud accounts.
", "votes": 16, "replies": [] }, { "id": 979291, "author_name": "inversion", "author_url": "", "post_date": "2020-08-20T18:38:56.657000", "content": "Update here:
\nhttps://www.kaggle.com/general/176169
", "votes": 14, "replies": [] }, { "id": 973624, "author_name": "Stanislav Blinov", "author_url": "", "post_date": "2020-08-17T12:07:02.310000", "content": "One thing I don't understand - WHY on Earth people pay 600-1000 $ for just a silver medal??
\nCorrect me if I'm wrong, but that's kinda a lot of money (well, maybe not if you're from US/EU) - just for example, that's brand new iPhone or a pretty solid PC or a week of vacation.
\nI remember someone wrote that it helps a lot when looking for jobs, but I seriously doubt it (especially in case of silver/bronze medals; it's understandable that gold may help in some way). Kaggle medals may be good in your resume but only if you already have some valuable experience. And, which matters most, if you're able to explain what you did to get those medals :D
TBH it also resembles buyers of loot and achievements in MMORPG/MOBA - just buying pixels to tickle one's sense of own importance.
", "votes": 13, "replies": [ { "id": 973756, "author_name": "Waylon Wu", "author_url": "", "post_date": "2020-08-17T13:50:28.507000", "content": "From the standpoint of a student, Kaggle medals are attractive. Student don't have any experience on projects and they don't know what's helpful to their job hunting. When they see a fancy ads that guarantees a job offer during their nervous job hunting period, it's likely that they take the shortcuts.
\nTheir kaggle path: get a medal -> learn -> (practice)
\nOur kaggle path: learn -> practice -> (get a medal)
Oh Kaggle medals are attractive for sure! I like Kaggle medals myself :D
\nTho I disagree with you - students have projects (course projects, scientific researches) and they know what's helpful (career fairs, networking, university advertisements for \"companies-partners\"). When I was in my grad year 2 years ago I already had 2 internships finished and some experience as translator and tech-writer. Is there really so much different situation in China?
\nI agree about shortcuts, but when one takes a shortcut, he should use his head and think how he will be using it. You buy stuff - you don't get to learn and practice - when you're asked about it - you're dead :)
Same for everywhere in the world. But you are talking about smart student, right? Apparently they are not. I can see they were not using their head and did not think from their $1000 charity.
", "votes": 1, "replies": [] }, { "id": 974771, "author_name": "", "author_url": "", "post_date": "2020-08-18T03:41:31.857000", "content": "There are many things in the world you can't fully understand if you don't put your feet in their shoes. Just like I can't understand why women are addicted to diamonds. It's obviously a fraud and IQ-tax collector, from my point of perspective though, but you have to pay it if you want to marry her. Women don't believe diamonds can guarantee their husbands' love are eternal, but they still can't help to buy diamonds….That's my metaphor of kaggle medals to those cheaters.
", "votes": 0, "replies": [] } ] }, { "id": 973622, "author_name": "Gunes Evitan", "author_url": "", "post_date": "2020-08-17T12:04:58.987000", "content": "Check these topics below.
\nhttps://www.kaggle.com/c/ashrae-energy-prediction/discussion/122203
\nhttps://www.kaggle.com/c/ieee-fraud-detection/discussion/110778
I wrote this post 8 months ago in the first topic and people instantly downvoted me and accused me of being racist.
\n\n\nHow do you explain other 100 novices from the same country in top 200 with very few submissions? Are their job also highly related to this competition?
\n
I still see the same pattern here. Some people get extremely defensive while trying to justify them.
", "votes": 13, "replies": [ { "id": 973623, "author_name": "Gilles Vandewiele", "author_url": "", "post_date": "2020-08-17T12:06:30.777000", "content": "Same for the Liverpool competition where @authman pointed out some very suspicious behaviour too.
", "votes": 2, "replies": [] }, { "id": 973628, "author_name": "Gunes Evitan", "author_url": "", "post_date": "2020-08-17T12:11:54.740000", "content": "Yes, I found that topic as well.
\nhttps://www.kaggle.com/c/liverpool-ion-switching/discussion/152417
\nGuess what? Someone was offended as usual.
", "votes": 4, "replies": [] }, { "id": 973633, "author_name": "Gilles Vandewiele", "author_url": "", "post_date": "2020-08-17T12:16:31.720000", "content": "I was expecting (and prepared for) those kind of comments, but I've been lucky to not yet have received such bullshit :)
", "votes": 5, "replies": [] }, { "id": 973646, "author_name": "903124", "author_url": "", "post_date": "2020-08-17T12:27:42.587000", "content": "\"数竞之家\" in ion switching submission is actually angtk.ai https://zhuanlan.zhihu.com/p/150559937
", "votes": 2, "replies": [] }, { "id": 973661, "author_name": "Gilles Vandewiele", "author_url": "", "post_date": "2020-08-17T12:44:25.583000", "content": "LOL, man stole my blog post hahaha
", "votes": 3, "replies": [] } ] }, { "id": 975292, "author_name": "earhian", "author_url": "", "post_date": "2020-08-18T08:27:28.700000", "content": "I think this is just the tip of the iceberg.
", "votes": 14, "replies": [ { "id": 975520, "author_name": "Yassine Alouini", "author_url": "", "post_date": "2020-08-18T10:33:33.687000", "content": "Hopefully, it is a small iceberg then. :D
", "votes": 1, "replies": [] }, { "id": 976923, "author_name": "from coffee import *", "author_url": "", "post_date": "2020-08-19T07:16:12.007000", "content": "You are probably right, the No-Code-Only-CSV-Submission-Competitions make sharing and cheating easy - and so many found it too intriguing to not try it.
\nFraud detection competition incoming ! ;)
", "votes": 2, "replies": [] } ] }, { "id": 974169, "author_name": "Innat", "author_url": "", "post_date": "2020-08-17T19:06:12.163000", "content": "@group16 Amazing catch. This is really important and needs immediate action.
\nThe suspected team member of yours that you've hinted, is he/she has 4/8 V100? 😂
\nI like to add one thing, in our multilingual toxic classification competition, we had a team of 5 members. At first, We competed alone but approximately 2 weeks before the deadline we're invited to join a team (4 members) around same scores.
The person who invited me was good in NLP and so we thought it would be great. After joining I found that out of four, 2 members have already gone completely silent. It was rest 3 of us. And guess what, they're from China (both). After that one has become competition expert without doing anything. After the competition, those users congratulated us when we posted our solution in the discussion. 😅
\nYour findings clarify many things to me. I always wonder that how can Joined 2/3/4/5/... days ago surpass many top kaggler (the real heroes). I don't want to repeat much, but now I can tell most of them are from China/Beijing. 😯
I hope private sharing doesn't break out like the corona epidemic :)
", "votes": 12, "replies": [ { "id": 973672, "author_name": "Vopani", "author_url": "", "post_date": "2020-08-17T12:56:57.643000", "content": "And the world waits for the Kaggle vaccine.
", "votes": 9, "replies": [] }, { "id": 974775, "author_name": "", "author_url": "", "post_date": "2020-08-18T03:45:14.033000", "content": "Well, this is like a type of virus and we don't have a vaccine yet…just like the coronavirus.
", "votes": 0, "replies": [] } ] }, { "id": 973664, "author_name": "Vopani", "author_url": "", "post_date": "2020-08-17T12:50:05.857000", "content": "Anything that has monetary gains will have scammers. I don't know how much of a priority Kaggle considers this to be but I won't be surprised if there is increasing amount of unfair practises in the whole system due to historic cases of cheaters getting away with it.
\nBut certainly if this continues it will be a shame that Competitions (also) is not clean anymore and it will be curtains down for many.
", "votes": 12, "replies": [ { "id": 973671, "author_name": "Aadhav Vignesh", "author_url": "", "post_date": "2020-08-17T12:56:05.393000", "content": "It would be really good if Kaggle comes up with a dedicated moderation team for competitions. This would lead to running of a smooth, fair competition and would help in resolving confusions created in competitions like Global Wheat Detection.
\nThe team would have the responsibilities to monitor solutions, remove non-compliant solutions, actively check for offenders and blacklist them.
\nKaggle users should also be required to verify themselves through a robust user verification system, which uses a bit more data (security experts would be better at explaining this).
", "votes": 4, "replies": [] } ] }, { "id": 974008, "author_name": "Heads or Tails", "author_url": "", "post_date": "2020-08-17T16:56:59.143000", "content": "@divyansh22 These kind of inappropriate and hateful comments are not welcome here. I trust that you will reflect and do better.
", "votes": 10, "replies": [ { "id": 974722, "author_name": "", "author_url": "", "post_date": "2020-08-18T03:03:33.593000", "content": "", "votes": 1, "replies": [] } ] }, { "id": 973997, "author_name": "Jie Lu", "author_url": "", "post_date": "2020-08-17T16:49:04.890000", "content": "@divyansh22 Wow always think twice before posting this kind of s**t thanks
", "votes": 10, "replies": [ { "id": 974716, "author_name": "", "author_url": "", "post_date": "2020-08-18T03:02:09.370000", "content": "", "votes": 1, "replies": [] } ] }, { "id": 973550, "author_name": "Changyi", "author_url": "", "post_date": "2020-08-17T11:08:48.277000", "content": "😲😲😲 jonney has changed name to jj-yytu and he's in the team just before me !
\nPlus, he WAS EJECTED in PANDA and in ALASKA2 (he was in silver in public and now inexist in LB I remember that because we were close).
\nThat means Kaggle knows he cheated but kept his account just because he's a MASTER ?
", "votes": 10, "replies": [ { "id": 973564, "author_name": "Changyi", "author_url": "", "post_date": "2020-08-17T11:19:08.463000", "content": "Now I understand how they can use 30+ submits to catch my team with 300+ submits.
", "votes": 4, "replies": [] }, { "id": 973634, "author_name": "inversion", "author_url": "", "post_date": "2020-08-17T12:16:34.287000", "content": "We have blocked the accounts of GMs in the past for cheating. When we have evidence of cheating, we take action. Sometimes, though, it takes time to collect and review all of the evidence.
", "votes": 26, "replies": [] }, { "id": 974296, "author_name": "Signal", "author_url": "", "post_date": "2020-08-17T21:17:12.267000", "content": "@vicioussong I hope you are wrong about that. It's hard to understand why Kaggle would allow someone to stay who was caught and even ejected in competitions. And then what? They see he continues to have newbies in his groups who contribute nothing, and they think anything has changed?
", "votes": 0, "replies": [] } ] }, { "id": 973486, "author_name": "Ahmet Erdem", "author_url": "", "post_date": "2020-08-17T10:23:48.790000", "content": "I don't understand who pays for Kaggle medals. I have never seen any benefit of having Kaggle medals.
", "votes": 11, "replies": [ { "id": 973491, "author_name": "Gilles Vandewiele", "author_url": "", "post_date": "2020-08-17T10:25:24.233000", "content": "It seems to help a lot when looking for jobs. Especially in larger companies where the hiring is done by HR people, whose technical skills are not sufficient to see through this scam. Also @aerdem4, aren't you working for a team that recruited people based on the fact that they are kaggle grandmasters? :P
", "votes": 14, "replies": [] }, { "id": 973500, "author_name": "brachester", "author_url": "", "post_date": "2020-08-17T10:31:41.633000", "content": "It can be used to pad your resume and is also a good talking point in interviews if you have limited professional experience.
", "votes": 0, "replies": [] }, { "id": 973516, "author_name": "Ahmet Erdem", "author_url": "", "post_date": "2020-08-17T10:45:22.037000", "content": "Unfortunately, I have opposite experience. I don't get my CV pass screening when I have Kaggle on it. And when recruiters call, they don't talk about Kaggle but work experience. They don't even know Kaggle.
\nNvidia and H2O are kind of unique in hiring Kagglers. I don't see other companies doing that. I am also against hiring people just because of their Kaggle title. You can become a grandmaster without solid knowledge in machine learning.
", "votes": 14, "replies": [] }, { "id": 973518, "author_name": "Ahmet Erdem", "author_url": "", "post_date": "2020-08-17T10:46:51.277000", "content": "Btw it is actually like buying a World of Warcraft character. You may buy the best character but you won't enjoy it since your skills don't match on that competitive level.
", "votes": 19, "replies": [] }, { "id": 973526, "author_name": "Gajendra Saraswat", "author_url": "", "post_date": "2020-08-17T10:50:17.940000", "content": "I remember that in this Melanoma competition, at the very starting, one guy from Japan commented that \"Please don't submit high scoring subs, otherwise I won't get any medal, in Japan, the companies prefer Kaggle medals.\"
", "votes": 0, "replies": [] }, { "id": 973533, "author_name": "brachester", "author_url": "", "post_date": "2020-08-17T10:54:18.160000", "content": "Some companies actually list Kaggle experience as a bonus in their listing. Twice interviewers have asked me about specific comps I attended and the techniques I used.
\nThat said, I doubt the medals worth as much as much as actual professional experience.
", "votes": 0, "replies": [] }, { "id": 973540, "author_name": "Ahmet Erdem", "author_url": "", "post_date": "2020-08-17T10:57:39.590000", "content": "I believe a Kaggle competition may worth even 1 year of work experience depending on the work and the competition. Especially in Europe, work usually progresses slow. But on the other hand, not all competitions are very fruitful, some of them are lottery.
", "votes": 2, "replies": [] }, { "id": 973544, "author_name": "Gajendra Saraswat", "author_url": "", "post_date": "2020-08-17T11:01:29.847000", "content": "Yeah, actually that's good that they focus on Kaggle competitions and ask the candidates about how they approached the competition because a competition here is about 3 months long, so it provides a interviewer the opportunity to ask many questions and ask thoroughly. Also, it is a huge boost to Kaggle's Image.
\nBut, if a company knows about such events, it will definitely malign the image of the community.
\nWhat we need now are some serious changes, so that no outsider can exploit the community and its culture.
", "votes": 0, "replies": [] }, { "id": 973548, "author_name": "brachester", "author_url": "", "post_date": "2020-08-17T11:06:54.873000", "content": "I mean I don't think any sensible person is going to hire you solely for your medal lol. It's just a way to get pass the non-technical HR. You still need to clear the technical interview regardless so I don't think those people that bought medal is going to benefit much from it.
", "votes": 3, "replies": [] }, { "id": 973570, "author_name": "Gajendra Saraswat", "author_url": "", "post_date": "2020-08-17T11:24:25.673000", "content": "Yeah, that's totally true!
", "votes": 0, "replies": [] }, { "id": 973597, "author_name": "Gunes Evitan", "author_url": "", "post_date": "2020-08-17T11:45:19.280000", "content": "If I remember correctly, I saw a post about the job market in China being very competitive. People invest lots of money to find an entry level job.
", "votes": 0, "replies": [] }, { "id": 973750, "author_name": "brachester", "author_url": "", "post_date": "2020-08-17T13:49:24.693000", "content": "I think it's competitive everywhere for entry level. New grads are a dime a dozen but companies (understandably) only want seniors.
", "votes": 0, "replies": [] } ] }, { "id": 974004, "author_name": "Aadhav Vignesh", "author_url": "", "post_date": "2020-08-17T16:54:13.117000", "content": "@divyansh22 Don't vent your anger on a specific country. Chinese competitors are not responsible for this small group of people. Your comment is totally racist. What has Kaggle to do COVID?
\n(Kaggle discussions seem to be broken. I can't reply to comments)
", "votes": 8, "replies": [ { "id": 974717, "author_name": "", "author_url": "", "post_date": "2020-08-18T03:02:44.700000", "content": "", "votes": 2, "replies": [] }, { "id": 974999, "author_name": "Aadhav Vignesh", "author_url": "", "post_date": "2020-08-18T05:35:31.603000", "content": "I'm glad that you accepted your fault. Hope you have a great day :)
", "votes": 1, "replies": [] } ] }, { "id": 974656, "author_name": "Mukharbek Organokov", "author_url": "", "post_date": "2020-08-18T02:19:42.583000", "content": "I think it's time to make at least ID and face verification as in EdX, at least. This will prevent such issues not 100%, but mostly.
\nAlso it will help good Kagglers not to team up someone wrong who can cheat and crash the work. Happened to my team at least once (proven one), which led to loss of TOP3% silver and 2.5 month of work.
", "votes": 7, "replies": [ { "id": 975526, "author_name": "Yassine Alouini", "author_url": "", "post_date": "2020-08-18T10:37:40.623000", "content": "That's a great idea: requiring some level of id would be a deterrent to a lot of \"fake\" accounts. I guess the technical implementation should take some point though. Or maybe it is something that Kaggle is already working on, we will see :)
", "votes": 0, "replies": [] } ] }, { "id": 973783, "author_name": "Raghawendra Singh", "author_url": "", "post_date": "2020-08-17T14:09:54.873000", "content": "What the hell is this???????? They are totally undermining the efforts of many Kagglers. Kaggle should take strong action and ban all those users who got any single medal using such platform.
", "votes": 7, "replies": [] }, { "id": 974246, "author_name": "Yassine Alouini", "author_url": "", "post_date": "2020-08-17T20:31:26.850000", "content": "Amazing investigative work! That's very bad for the Kaggle community and makes the ranking a little less meaningful, which is sad. ☹️
\nThat being said, nothing beats the joy of learning something new and exploring new ideas, so I guess these that pay for medals won't get that. 😄
\nFinally, if I may ask, do you know why you were contacted? Did it happen after you teamed up with the suspicious team-mate or was that just a coincidence?
\nAgain, thanks for the post!
", "votes": 8, "replies": [ { "id": 974248, "author_name": "Gilles Vandewiele", "author_url": "", "post_date": "2020-08-17T20:33:42.540000", "content": "Probably due to my previous post :)
", "votes": 1, "replies": [ { "id": 974260, "author_name": "Signal", "author_url": "", "post_date": "2020-08-17T20:48:46.097000", "content": "It wouldn't be necessarily, a bad idea, to offer some sort of amnesty. Those who come forward, confess, are not sanctioned and not doxed. Perhaps their medals are stripped…….I mean, there are consequences, but if they are a real person, not a bot/mule/farm account, then they can be given a clean slate going forward and they would be doing us all a big help, by turning in as much info as possible.
", "votes": 2, "replies": [] } ] }, { "id": 974252, "author_name": "Yassine Alouini", "author_url": "", "post_date": "2020-08-17T20:37:02.993000", "content": "Will give it a look, cheers!
", "votes": 0, "replies": [] } ] }, { "id": 974221, "author_name": "alijs", "author_url": "", "post_date": "2020-08-17T20:00:14.713000", "content": "The value of things is determined by how easy or hard is to get them. If it will be possible to simply buy Kaggle medals, their value will decrease. So we all, including Kaggle and competitors, are interested that cheaters are revealed and their medals removed.
", "votes": 8, "replies": [] }, { "id": 974115, "author_name": "Signal", "author_url": "", "post_date": "2020-08-17T18:16:27.550000", "content": "@inversion plenty of evidence here and also here https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/174635#973667 .
\nedu.angtk.com took their website down as @group16 , I and others were challenging them last night. You see the main representative https://www.kaggle.com/lifeifei38 , who apparently is part of the staff for this medal farm, has an account a few months old with no activity. You can't let them get away with these medals in Wheat and Melanoma……….the lid is blown open on this group, and they have been doing this apparently for at least 2 years, but it's blatant, \"in your face\" now. I mean just look at the groups with user @yikououbao . Thats pure medal farming and yikououbao has taken real medals, and prevented many others from getting them.
", "votes": 8, "replies": [] }, { "id": 973647, "author_name": "FChmiel", "author_url": "", "post_date": "2020-08-17T12:27:53.707000", "content": "I think the danger of this situation cannot be stressed enough: It threatens to undermine the community based on the open-source premise which makes it fair for all to compete, regardless of sociodemographic background. Open, and transparent collaboration strengthens the community by building trust in the methods and helping to push the cutting edge.
\nCompanies / methods like this are toxic and damage the progress made by the wider open source community and the ideals they champion. Taking such large sums of money off people who may not know better is a low-grade business tactic - shame on them.
\n$500 is a lot of money even for someone from the US/EU! For comparison, High-caliber ML courses vary from free to tens of dollars…
", "votes": 8, "replies": [] }, { "id": 973492, "author_name": "FChmiel", "author_url": "", "post_date": "2020-08-17T10:28:52.630000", "content": "The account you mention looks highly likely to be a medal farm that tricks its way onto high-scoring teams. I expect the photo was chosen to increase credibility and to influence people.
", "votes": 8, "replies": [ { "id": 973499, "author_name": "Gilles Vandewiele", "author_url": "", "post_date": "2020-08-17T10:31:23.530000", "content": "My thoughts exactly. I just hope that it is only that and that the account is in no way involved with the private sharing. I'll gladly help the people from Kaggle in any way possible to further clear this out.
", "votes": 2, "replies": [] }, { "id": 974749, "author_name": "", "author_url": "", "post_date": "2020-08-18T03:25:25.193000", "content": "Very typical \"successful customer story\" tricks in the marketing domain.
", "votes": 0, "replies": [] } ] }, { "id": 974140, "author_name": "Heads or Tails", "author_url": "", "post_date": "2020-08-17T18:38:39.130000", "content": "As a side note, and for what it's worth, I feel like it would be perfectly fine for the private LB validation to take more time in general. This doesn't have to be an instantaneous switch; especially if the additional time helps to make sure that all the final placings are legit. As long as the time scales are clearly communicated, some delay might even help in keeping more people interested in post-submission deadline discussions and solution sharing.
", "votes": 5, "replies": [ { "id": 974148, "author_name": "Stanislav Blinov", "author_url": "", "post_date": "2020-08-17T18:44:09.203000", "content": "It's OK to take SOME more time instead of instant switching, but we don't need another ASHRAE.
", "votes": 1, "replies": [] } ] }, { "id": 973730, "author_name": "", "author_url": "", "post_date": "2020-08-17T13:34:57.720000", "content": "", "votes": 5, "replies": [] }, { "id": 973630, "author_name": "", "author_url": "", "post_date": "2020-08-17T12:14:13.623000", "content": "", "votes": 5, "replies": [ { "id": 974315, "author_name": "", "author_url": "", "post_date": "2020-08-17T22:01:20.370000", "content": "", "votes": 0, "replies": [] } ] }, { "id": 973592, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:41:36.763000", "content": "", "votes": 5, "replies": [ { "id": 973626, "author_name": "", "author_url": "", "post_date": "2020-08-17T12:11:13.247000", "content": "", "votes": 1, "replies": [] }, { "id": 974761, "author_name": "", "author_url": "", "post_date": "2020-08-18T03:33:33.887000", "content": "", "votes": 0, "replies": [] } ] }, { "id": 976179, "author_name": "", "author_url": "", "post_date": "2020-08-18T17:38:09.773000", "content": "", "votes": 6, "replies": [ { "id": 976183, "author_name": "", "author_url": "", "post_date": "2020-08-18T17:41:55.480000", "content": "", "votes": 0, "replies": [] }, { "id": 976185, "author_name": "", "author_url": "", "post_date": "2020-08-18T17:43:51.427000", "content": "", "votes": 6, "replies": [] }, { "id": 976188, "author_name": "", "author_url": "", "post_date": "2020-08-18T17:45:19.920000", "content": "", "votes": 2, "replies": [] }, { "id": 976672, "author_name": "", "author_url": "", "post_date": "2020-08-19T03:03:11.867000", "content": "", "votes": 1, "replies": [] }, { "id": 976675, "author_name": "", "author_url": "", "post_date": "2020-08-19T03:06:05.627000", "content": "", "votes": 1, "replies": [] } ] }, { "id": 974310, "author_name": "", "author_url": "", "post_date": "2020-08-17T21:34:14.223000", "content": "", "votes": 4, "replies": [ { "id": 974346, "author_name": "", "author_url": "", "post_date": "2020-08-17T23:12:44.240000", "content": "", "votes": 5, "replies": [ { "id": 974468, "author_name": "", "author_url": "", "post_date": "2020-08-18T00:36:37.473000", "content": "", "votes": 2, "replies": [] }, { "id": 974677, "author_name": "", "author_url": "", "post_date": "2020-08-18T02:29:12.017000", "content": "", "votes": 0, "replies": [] }, { "id": 974686, "author_name": "", "author_url": "", "post_date": "2020-08-18T02:35:05.120000", "content": "", "votes": 2, "replies": [] }, { "id": 974689, "author_name": "", "author_url": "", "post_date": "2020-08-18T02:38:17.430000", "content": "", "votes": 1, "replies": [] }, { "id": 974696, "author_name": "", "author_url": "", "post_date": "2020-08-18T02:39:15.433000", "content": "", "votes": 1, "replies": [] } ] } ] }, { "id": 973509, "author_name": "", "author_url": "", "post_date": "2020-08-17T10:38:26.523000", "content": "", "votes": 4, "replies": [] }, { "id": 973493, "author_name": "", "author_url": "", "post_date": "2020-08-17T10:29:32.680000", "content": "", "votes": 5, "replies": [] }, { "id": 974883, "author_name": "", "author_url": "", "post_date": "2020-08-18T04:40:37.007000", "content": "", "votes": 3, "replies": [] }, { "id": 979044, "author_name": "", "author_url": "", "post_date": "2020-08-20T15:25:55.123000", "content": "", "votes": 3, "replies": [ { "id": 979152, "author_name": "", "author_url": "", "post_date": "2020-08-20T16:55:18.093000", "content": "", "votes": 2, "replies": [] }, { "id": 979205, "author_name": "", "author_url": "", "post_date": "2020-08-20T17:33:03.330000", "content": "", "votes": 4, "replies": [] }, { "id": 979221, "author_name": "", "author_url": "", "post_date": "2020-08-20T17:43:45.963000", "content": "", "votes": 1, "replies": [] }, { "id": 979225, "author_name": "", "author_url": "", "post_date": "2020-08-20T17:44:45.923000", "content": "", "votes": 4, "replies": [] }, { "id": 979235, "author_name": "", "author_url": "", "post_date": "2020-08-20T17:51:08.367000", "content": "", "votes": 4, "replies": [] }, { "id": 979244, "author_name": "", "author_url": "", "post_date": "2020-08-20T17:56:48.150000", "content": "", "votes": 1, "replies": [] }, { "id": 979282, "author_name": "", "author_url": "", "post_date": "2020-08-20T18:29:01.877000", "content": "", "votes": 5, "replies": [] }, { "id": 979328, "author_name": "", "author_url": "", "post_date": "2020-08-20T19:04:43.237000", "content": "", "votes": 4, "replies": [] } ] }, { "id": 975476, "author_name": "", "author_url": "", "post_date": "2020-08-18T10:05:41.140000", "content": "", "votes": 3, "replies": [] }, { "id": 975122, "author_name": "", "author_url": "", "post_date": "2020-08-18T06:53:18.167000", "content": "", "votes": 3, "replies": [] }, { "id": 974661, "author_name": "", "author_url": "", "post_date": "2020-08-18T02:21:49.850000", "content": "", "votes": 3, "replies": [ { "id": 975524, "author_name": "", "author_url": "", "post_date": "2020-08-18T10:36:05.137000", "content": "", "votes": 0, "replies": [] }, { "id": 975529, "author_name": "", "author_url": "", "post_date": "2020-08-18T10:41:41.087000", "content": "", "votes": 3, "replies": [] }, { "id": 975541, "author_name": "", "author_url": "", "post_date": "2020-08-18T10:48:45.797000", "content": "", "votes": 3, "replies": [] }, { "id": 975548, "author_name": "", "author_url": "", "post_date": "2020-08-18T10:54:03.577000", "content": "", "votes": 3, "replies": [] }, { "id": 975559, "author_name": "", "author_url": "", "post_date": "2020-08-18T10:57:48.803000", "content": "", "votes": 1, "replies": [] } ] }, { "id": 974100, "author_name": "", "author_url": "", "post_date": "2020-08-17T18:12:54.490000", "content": "", "votes": 3, "replies": [ { "id": 974145, "author_name": "", "author_url": "", "post_date": "2020-08-17T18:42:06.007000", "content": "", "votes": 2, "replies": [] }, { "id": 974157, "author_name": "", "author_url": "", "post_date": "2020-08-17T18:55:06.233000", "content": "", "votes": 2, "replies": [] }, { "id": 974166, "author_name": "", "author_url": "", "post_date": "2020-08-17T19:04:00.750000", "content": "", "votes": 1, "replies": [] }, { "id": 974217, "author_name": "", "author_url": "", "post_date": "2020-08-17T19:52:09.637000", "content": "", "votes": 0, "replies": [] } ] }, { "id": 974099, "author_name": "", "author_url": "", "post_date": "2020-08-17T18:12:29.520000", "content": "", "votes": 3, "replies": [] }, { "id": 974029, "author_name": "", "author_url": "", "post_date": "2020-08-17T17:15:56.847000", "content": "", "votes": 3, "replies": [] }, { "id": 973940, "author_name": "", "author_url": "", "post_date": "2020-08-17T16:10:10.300000", "content": "", "votes": 3, "replies": [] }, { "id": 976771, "author_name": "", "author_url": "", "post_date": "2020-08-19T04:48:47.133000", "content": "", "votes": 4, "replies": [ { "id": 976817, "author_name": "", "author_url": "", "post_date": "2020-08-19T05:39:13.633000", "content": "", "votes": 6, "replies": [] }, { "id": 976825, "author_name": "", "author_url": "", "post_date": "2020-08-19T05:47:39.073000", "content": "", "votes": 1, "replies": [] }, { "id": 976865, "author_name": "", "author_url": "", "post_date": "2020-08-19T06:30:22.180000", "content": "", "votes": 0, "replies": [] }, { "id": 976919, "author_name": "", "author_url": "", "post_date": "2020-08-19T07:12:53.413000", "content": "", "votes": 2, "replies": [] }, { "id": 976929, "author_name": "", "author_url": "", "post_date": "2020-08-19T07:21:03.923000", "content": "", "votes": 3, "replies": [] }, { "id": 977338, "author_name": "", "author_url": "", "post_date": "2020-08-19T12:30:04.957000", "content": "", "votes": 1, "replies": [] }, { "id": 977395, "author_name": "", "author_url": "", "post_date": "2020-08-19T13:01:29.160000", "content": "", "votes": 2, "replies": [] }, { "id": 977459, "author_name": "", "author_url": "", "post_date": "2020-08-19T13:42:46.617000", "content": "", "votes": 0, "replies": [] }, { "id": 981257, "author_name": "", "author_url": "", "post_date": "2020-08-22T10:26:15.393000", "content": "", "votes": 1, "replies": [] }, { "id": 981276, "author_name": "", "author_url": "", "post_date": "2020-08-22T10:40:07.880000", "content": "", "votes": 0, "replies": [] }, { "id": 981281, "author_name": "", "author_url": "", "post_date": "2020-08-22T10:44:33.677000", "content": "", "votes": 0, "replies": [] }, { "id": 981290, "author_name": "", "author_url": "", "post_date": "2020-08-22T10:50:23.337000", "content": "", "votes": 1, "replies": [] }, { "id": 981309, "author_name": "", "author_url": "", "post_date": "2020-08-22T11:09:46.100000", "content": "", "votes": 0, "replies": [] }, { "id": 982006, "author_name": "", "author_url": "", "post_date": "2020-08-23T00:13:26.323000", "content": "", "votes": 0, "replies": [] } ] }, { "id": 975483, "author_name": "", "author_url": "", "post_date": "2020-08-18T10:10:38.483000", "content": "", "votes": 4, "replies": [ { "id": 975518, "author_name": "", "author_url": "", "post_date": "2020-08-18T10:32:44.733000", "content": "", "votes": 2, "replies": [] }, { "id": 975614, "author_name": "", "author_url": "", "post_date": "2020-08-18T11:35:20.263000", "content": "", "votes": 1, "replies": [] }, { "id": 975638, "author_name": "", "author_url": "", "post_date": "2020-08-18T11:48:17.890000", "content": "", "votes": 0, "replies": [] }, { "id": 975654, "author_name": "", "author_url": "", "post_date": "2020-08-18T12:03:13.133000", "content": "", "votes": 2, "replies": [] }, { "id": 976199, "author_name": "", "author_url": "", "post_date": "2020-08-18T17:51:22.397000", "content": "", "votes": 0, "replies": [] }, { "id": 976678, "author_name": "", "author_url": "", "post_date": "2020-08-19T03:12:58.480000", "content": "", "votes": 1, "replies": [] } ] }, { "id": 974014, "author_name": "", "author_url": "", "post_date": "2020-08-17T17:00:06.280000", "content": "", "votes": 4, "replies": [] }, { "id": 973941, "author_name": "", "author_url": "", "post_date": "2020-08-17T16:10:16.683000", "content": "", "votes": 4, "replies": [] }, { "id": 973675, "author_name": "", "author_url": "", "post_date": "2020-08-17T12:58:20.060000", "content": "", "votes": 4, "replies": [] }, { "id": 973639, "author_name": "", "author_url": "", "post_date": "2020-08-17T12:20:24.833000", "content": "", "votes": 4, "replies": [ { "id": 973642, "author_name": "", "author_url": "", "post_date": "2020-08-17T12:22:59.713000", "content": "", "votes": 0, "replies": [] } ] }, { "id": 973478, "author_name": "", "author_url": "", "post_date": "2020-08-17T10:20:20.677000", "content": "", "votes": 4, "replies": [] }, { "id": 975931, "author_name": "", "author_url": "", "post_date": "2020-08-18T14:37:25.783000", "content": "", "votes": 2, "replies": [ { "id": 976353, "author_name": "", "author_url": "", "post_date": "2020-08-18T19:59:41.103000", "content": "", "votes": 11, "replies": [] }, { "id": 976586, "author_name": "", "author_url": "", "post_date": "2020-08-19T01:01:12.183000", "content": "", "votes": 1, "replies": [] } ] }, { "id": 974060, "author_name": "", "author_url": "", "post_date": "2020-08-17T17:42:02.167000", "content": "", "votes": 2, "replies": [ { "id": 974735, "author_name": "", "author_url": "", "post_date": "2020-08-18T03:15:53.203000", "content": "", "votes": 1, "replies": [] } ] }, { "id": 973937, "author_name": "", "author_url": "", "post_date": "2020-08-17T16:08:00.047000", "content": "", "votes": 2, "replies": [] }, { "id": 973932, "author_name": "", "author_url": "", "post_date": "2020-08-17T16:02:13.940000", "content": "", "votes": 3, "replies": [] }, { "id": 973545, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:02:05.230000", "content": "", "votes": 3, "replies": [ { "id": 973571, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:26:10.833000", "content": "", "votes": -3, "replies": [] } ] }, { "id": 974056, "author_name": "", "author_url": "", "post_date": "2020-08-17T17:40:19.807000", "content": "", "votes": -1, "replies": [ { "id": 974126, "author_name": "", "author_url": "", "post_date": "2020-08-17T18:22:42.037000", "content": "", "votes": -3, "replies": [] }, { "id": 974170, "author_name": "", "author_url": "", "post_date": "2020-08-17T19:07:10.173000", "content": "", "votes": 10, "replies": [] }, { "id": 974733, "author_name": "", "author_url": "", "post_date": "2020-08-18T03:13:11.847000", "content": "", "votes": 2, "replies": [] } ] }, { "id": 980592, "author_name": "", "author_url": "", "post_date": "2020-08-21T17:55:45.083000", "content": "", "votes": 1, "replies": [] }, { "id": 975542, "author_name": "", "author_url": "", "post_date": "2020-08-18T10:49:25.173000", "content": "", "votes": 1, "replies": [ { "id": 975579, "author_name": "", "author_url": "", "post_date": "2020-08-18T11:13:14.597000", "content": "", "votes": 0, "replies": [] }, { "id": 976344, "author_name": "", "author_url": "", "post_date": "2020-08-18T19:54:15.557000", "content": "", "votes": 1, "replies": [] } ] }, { "id": 974142, "author_name": "", "author_url": "", "post_date": "2020-08-17T18:38:51.080000", "content": "", "votes": 1, "replies": [] }, { "id": 974044, "author_name": "", "author_url": "", "post_date": "2020-08-17T17:33:35.553000", "content": "", "votes": 1, "replies": [] }, { "id": 974011, "author_name": "", "author_url": "", "post_date": "2020-08-17T16:59:32.657000", "content": "", "votes": 1, "replies": [] }, { "id": 973890, "author_name": "", "author_url": "", "post_date": "2020-08-17T15:30:21.447000", "content": "", "votes": 1, "replies": [] }, { "id": 973874, "author_name": "", "author_url": "", "post_date": "2020-08-17T15:17:32.690000", "content": "", "votes": 1, "replies": [] }, { "id": 973843, "author_name": "", "author_url": "", "post_date": "2020-08-17T14:54:30.960000", "content": "", "votes": 1, "replies": [] }, { "id": 973794, "author_name": "", "author_url": "", "post_date": "2020-08-17T14:16:50.397000", "content": "", "votes": 1, "replies": [] }, { "id": 973681, "author_name": "", "author_url": "", "post_date": "2020-08-17T13:04:34.467000", "content": "", "votes": 1, "replies": [] }, { "id": 973679, "author_name": "", "author_url": "", "post_date": "2020-08-17T13:03:17.817000", "content": "", "votes": 1, "replies": [] }, { "id": 973618, "author_name": "", "author_url": "", "post_date": "2020-08-17T12:02:09.610000", "content": "", "votes": 1, "replies": [] }, { "id": 973598, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:46:30.480000", "content": "", "votes": 1, "replies": [] }, { "id": 973523, "author_name": "", "author_url": "", "post_date": "2020-08-17T10:48:25.127000", "content": "", "votes": 1, "replies": [ { "id": 975922, "author_name": "", "author_url": "", "post_date": "2020-08-18T14:24:37.427000", "content": "", "votes": 0, "replies": [] } ] }, { "id": 974789, "author_name": "", "author_url": "", "post_date": "2020-08-18T03:53:43.770000", "content": "", "votes": 2, "replies": [] }, { "id": 974514, "author_name": "", "author_url": "", "post_date": "2020-08-18T01:04:30.347000", "content": "", "votes": 2, "replies": [ { "id": 974527, "author_name": "", "author_url": "", "post_date": "2020-08-18T01:12:58.160000", "content": "", "votes": 0, "replies": [] }, { "id": 974530, "author_name": "", "author_url": "", "post_date": "2020-08-18T01:15:07.360000", "content": "", "votes": 2, "replies": [] } ] }, { "id": 974081, "author_name": "", "author_url": "", "post_date": "2020-08-17T18:00:41.037000", "content": "", "votes": 2, "replies": [] }, { "id": 973989, "author_name": "", "author_url": "", "post_date": "2020-08-17T16:46:05.763000", "content": "", "votes": 2, "replies": [] }, { "id": 973945, "author_name": "", "author_url": "", "post_date": "2020-08-17T16:12:44.720000", "content": "", "votes": 0, "replies": [] }, { "id": 974070, "author_name": "", "author_url": "", "post_date": "2020-08-17T17:52:17.743000", "content": "", "votes": 1, "replies": [ { "id": 974146, "author_name": "", "author_url": "", "post_date": "2020-08-17T18:42:43.217000", "content": "", "votes": 2, "replies": [] }, { "id": 974794, "author_name": "", "author_url": "", "post_date": "2020-08-18T03:56:27.630000", "content": "", "votes": 0, "replies": [] } ] }, { "id": 978980, "author_name": "", "author_url": "", "post_date": "2020-08-20T14:50:29.133000", "content": "", "votes": -5, "replies": [] }, { "id": 979201, "author_name": "", "author_url": "", "post_date": "2020-08-20T17:29:07.370000", "content": "", "votes": -4, "replies": [] }, { "id": 978911, "author_name": "", "author_url": "", "post_date": "2020-08-20T13:47:22.027000", "content": "", "votes": -6, "replies": [] }, { "id": 987357, "author_name": "", "author_url": "", "post_date": "2020-08-27T07:29:40.653000", "content": "", "votes": 0, "replies": [ { "id": 987390, "author_name": "", "author_url": "", "post_date": "2020-08-27T08:12:04.360000", "content": "", "votes": 0, "replies": [] } ] }, { "id": 986630, "author_name": "", "author_url": "", "post_date": "2020-08-26T16:33:05.913000", "content": "", "votes": 0, "replies": [] }, { "id": 985889, "author_name": "", "author_url": "", "post_date": "2020-08-26T04:55:43.120000", "content": "", "votes": 0, "replies": [] }, { "id": 981997, "author_name": "", "author_url": "", "post_date": "2020-08-22T23:55:29.303000", "content": "", "votes": 0, "replies": [] }, { "id": 979068, "author_name": "", "author_url": "", "post_date": "2020-08-20T15:40:46.723000", "content": "", "votes": 0, "replies": [] }, { "id": 975895, "author_name": "", "author_url": "", "post_date": "2020-08-18T14:09:57.067000", "content": "", "votes": 0, "replies": [] }, { "id": 974367, "author_name": "", "author_url": "", "post_date": "2020-08-17T23:48:14.737000", "content": "", "votes": 0, "replies": [] }, { "id": 974266, "author_name": "", "author_url": "", "post_date": "2020-08-17T20:55:35.673000", "content": "", "votes": 0, "replies": [ { "id": 974274, "author_name": "", "author_url": "", "post_date": "2020-08-17T21:02:17.507000", "content": "", "votes": 1, "replies": [] }, { "id": 976932, "author_name": "", "author_url": "", "post_date": "2020-08-19T07:22:11.617000", "content": "", "votes": 0, "replies": [] } ] }, { "id": 974212, "author_name": "", "author_url": "", "post_date": "2020-08-17T19:48:20.817000", "content": "", "votes": 0, "replies": [] }, { "id": 974085, "author_name": "", "author_url": "", "post_date": "2020-08-17T18:02:11.370000", "content": "", "votes": 0, "replies": [] }, { "id": 974020, "author_name": "", "author_url": "", "post_date": "2020-08-17T17:04:29.267000", "content": "", "votes": 0, "replies": [] }, { "id": 973947, "author_name": "", "author_url": "", "post_date": "2020-08-17T16:14:03.370000", "content": "", "votes": 0, "replies": [] }, { "id": 973568, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:23:11.843000", "content": "", "votes": 0, "replies": [ { "id": 973574, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:27:59.503000", "content": "", "votes": 19, "replies": [ { "id": 973611, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:57:33.840000", "content": "", "votes": 1, "replies": [] }, { "id": 973621, "author_name": "", "author_url": "", "post_date": "2020-08-17T12:04:36.083000", "content": "", "votes": 5, "replies": [] }, { "id": 973625, "author_name": "", "author_url": "", "post_date": "2020-08-17T12:08:50.613000", "content": "", "votes": 3, "replies": [] } ] }, { "id": 973581, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:33:36.527000", "content": "", "votes": 1, "replies": [] }, { "id": 973585, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:38:28.113000", "content": "", "votes": 4, "replies": [] } ] }, { "id": 981181, "author_name": "", "author_url": "", "post_date": "2020-08-22T08:46:17", "content": "", "votes": -2, "replies": [] }, { "id": 979451, "author_name": "", "author_url": "", "post_date": "2020-08-20T21:09:26.007000", "content": "", "votes": 0, "replies": [] }, { "id": 976904, "author_name": "", "author_url": "", "post_date": "2020-08-19T07:04:32.097000", "content": "", "votes": -1, "replies": [ { "id": 976907, "author_name": "", "author_url": "", "post_date": "2020-08-19T07:07:07.460000", "content": "", "votes": 5, "replies": [] } ] }, { "id": 976901, "author_name": "", "author_url": "", "post_date": "2020-08-19T07:00:46.080000", "content": "", "votes": 2, "replies": [] }, { "id": 975918, "author_name": "", "author_url": "", "post_date": "2020-08-18T14:22:26.090000", "content": "", "votes": 0, "replies": [] }, { "id": 974403, "author_name": "", "author_url": "", "post_date": "2020-08-18T00:12:48.370000", "content": "", "votes": 0, "replies": [ { "id": 977740, "author_name": "", "author_url": "", "post_date": "2020-08-19T17:20:05.037000", "content": "", "votes": 1, "replies": [] } ] }, { "id": 974009, "author_name": "", "author_url": "", "post_date": "2020-08-17T16:57:42.250000", "content": "", "votes": 0, "replies": [] }, { "id": 973761, "author_name": "", "author_url": "", "post_date": "2020-08-17T13:55:48.067000", "content": "", "votes": -5, "replies": [ { "id": 973788, "author_name": "", "author_url": "", "post_date": "2020-08-17T14:13:42.577000", "content": "", "votes": 11, "replies": [] } ] }, { "id": 973678, "author_name": "", "author_url": "", "post_date": "2020-08-17T13:02:53.253000", "content": "", "votes": 0, "replies": [] }, { "id": 973677, "author_name": "", "author_url": "", "post_date": "2020-08-17T13:02:22.380000", "content": "", "votes": 0, "replies": [] }, { "id": 973578, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:29:58.783000", "content": "", "votes": -5, "replies": [ { "id": 973582, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:34:36.163000", "content": "", "votes": 2, "replies": [] }, { "id": 973583, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:36:54.203000", "content": "", "votes": -1, "replies": [] }, { "id": 973584, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:36:55.500000", "content": "", "votes": 2, "replies": [] }, { "id": 973586, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:39:18.970000", "content": "", "votes": -1, "replies": [] }, { "id": 973588, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:40:08.020000", "content": "", "votes": 1, "replies": [] }, { "id": 973591, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:40:48.457000", "content": "", "votes": 1, "replies": [] }, { "id": 973594, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:43:41.450000", "content": "", "votes": -2, "replies": [] }, { "id": 973600, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:48:14.690000", "content": "", "votes": 1, "replies": [] }, { "id": 973602, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:50:38.783000", "content": "", "votes": 2, "replies": [] }, { "id": 973603, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:51:05.917000", "content": "", "votes": 0, "replies": [] }, { "id": 973607, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:53:00.717000", "content": "", "votes": 3, "replies": [] }, { "id": 973610, "author_name": "", "author_url": "", "post_date": "2020-08-17T11:54:46.483000", "content": "", "votes": 1, "replies": [] } ] }, { "id": 976704, "author_name": "", "author_url": "", "post_date": "2020-08-19T03:48:05.437000", "content": "", "votes": -4, "replies": [] } ], "raw_markdown_by_id": { "973472": "I got contacted by a source that prefers to stay anonymous with some very valuable information that I would like to make public.\n\n**Why are people private sharing?**\nOne could ask why people share their notebooks or submission files and decrease their own odds of winning. The answer is simple: money. There are organizations out there that sell Kaggle medals for quite a large amount of money. Here's an [example of such a store](https://shop319532120.world.taobao.com/?spm=a230r.7195193.1997079397.2.52a85a59xgPAS6). Here, a guaranteed 1% finish (silver medal) is being sold for 1000 yuan (apparently that has to be paid 4 times which results in 600 USD).\n\n\n\nHere's another screenshot where they boast with how 2 of their students got to master tier and how they secured 38 (!!) silver medals within one competition. What a success rate!\n\n\nHere's a rough outline of how the process goes:\n1. students find them on websites or shops(taobao淘宝)\n2. they talk to the front desk about what's going on, asking for prices and services included\n3. they have different prices for different medals, for instance, if you want a silver, you need to pay more\n4. their lessons do not provide anything meaningful to anyone, using it as a mask for their true business\n5. you will also be assigned to a teacher where he answers your questions really really poorly.\n6. in kernel only competitions, they will give you models. In csv competitions, they will give you subs(for instance in SIIM)\n7. after you finish this competition, you learned nothing but how to cheat\n\n**Who is behind this?**\nAdvertisements that include the names of two kaggle masters can be found:\n\n\n\n\n\nThese are screenshots from different competitions. What a coincidence, but these two accounts are currently ranked 18th on the public LB as well. Not too far away from all the fake accounts on the top. The chinese name of the first master is 王苏宏, which is also used to advertise 20 different accounts in the deepfake competition:\n\n\n**What else can we do?**\nWe need to identify all accounts related to this suspicious cluster. One way to do this would be to create a graph analysis based on the [Kaggle meta dataset](https://www.kaggle.com/kaggle/meta-kaggle).\n\n**I would also like to clarify that I might have blood on my hands. During this competition, I teamed up with someone that seemed very trust-worthy at first glance based on previous competition results & current LB standing. After teaming up, this person went completely silent and we never heard from him/her again. We also suspect the profile picture used to be fake (as it can be found on other instances with Google Images). I'll clarify this further after the competition to avoid any kind of griefing.**", "973631": "Almost all the golden area is overtaken by novice cheaters accounts. \n\nIf they're still there on private LB, I will likely never compete again on kaggle. ", "973698": "I hope Kaggle will ban all of those who buy or sell medals.", "974218": "You have a future in Intelligence industry ;)", "973849": "Thank you, Gilles and your anonymous source, for exposing those fraudulent schemes. These methods constitute such a blatant disrespect of the community; it is very disappointing to see. I hope that the Kaggle team will take swift and decisive actions against them. The hard work and effort of all honest competitors should not be undermined like this.\n\nTo Kaggle's credit, we know that accounts of cheaters are regularly removed, and that a lot of work is being done behind the scenes to keep the community clean. In the past, even cheating \"Grandmasters\" have been banned. However, those higher-profile cases normally take quite a lot of time, during which the bad actors can get away with more cheating. While I can understand that Kaggle wants to err on the side of caution, time is really of the essence here. A lot of damage is being done if groups of people can get away with fraud for even 1 competition, since that invites copycat behaviour.\n\n**As Kaggle continues to grow in popularity, anti-fraud actions need to be taken faster. Otherwise our community might gradually lose its hard-earned reputation.**\n\nI would also like to say, Gilles, that you and your honest teammates did nothing wrong during this competition. You did your due diligence before teaming up, and in a better situation this shady account would likely have never had the results to make it seem trustworthy. I hope that Kaggle will find a way to punish the cheaters without it affecting the honest competitors.", "974013": "@divyansh22 if I see a cheater from your country, I would never say your country is full of cheaters.\n\nPlease pay attention, we're talking about cheating not racism.", "973497": "Fortunately I did not delete my WeChat app after Kaggle Days in Beijing. Let's go for shopping!\n\nBtw I also have 11 silver medals and a few of them I don't need anymore if anyone interested :D", "973790": "For those who’ve worked in the fraud detection space, it’s so interesting because transaction time (or submission time in our case) is such an interesting feature. Intuitively, you’d think they’d stagger their submissions to make the fraud seem more organic, but that goes against another driving factor, which is the desire to make as much money off the scheme as quickly as possible before the lid gets blown off the operation.\n\n🍿🍿🍿🍿🍿", "973736": "Nice work, @group16 . Kaggle needs you, the world needs you. :D", "974159": "On a lighter note, i think it's time we will soon see a kaggle competition about detecting fraud accounts.", "979291": "Update here:\n\nhttps://www.kaggle.com/general/176169", "973624": "One thing I don't understand - WHY on Earth people pay 600-1000 $ for just a silver medal??\nCorrect me if I'm wrong, but that's kinda a lot of money (well, maybe not if you're from US/EU) - just for example, that's brand new iPhone or a pretty solid PC or a week of vacation.\nI remember someone wrote that it helps a lot when looking for jobs, but I seriously doubt it (especially in case of silver/bronze medals; it's understandable that gold may help in some way). Kaggle medals may be good in your resume but only if you already have some valuable experience. And, which matters most, if you're able to explain what you did to get those medals :D\n\nTBH it also resembles buyers of loot and achievements in MMORPG/MOBA - just buying pixels to tickle one's sense of own importance.", "973622": "Check these topics below.\nhttps://www.kaggle.com/c/ashrae-energy-prediction/discussion/122203\nhttps://www.kaggle.com/c/ieee-fraud-detection/discussion/110778\n\nI wrote this post 8 months ago in the first topic and people instantly downvoted me and accused me of being racist.\n\n> How do you explain other 100 novices from the same country in top 200 with very few submissions? Are their job also highly related to this competition?\n\nI still see the same pattern here. Some people get extremely defensive while trying to justify them.\n\n\n", "975292": "I think this is just the tip of the iceberg. ", "974169": "@group16 Amazing catch. This is really important and needs immediate action. \n\nThe suspected team member of yours that you've hinted, is he/she has 4/8 V100? 😂\nI like to add one thing, in our multilingual toxic classification competition, we had a team of 5 members. At first, We competed alone but approximately 2 weeks before the deadline we're invited to join a team (4 members) around same scores. \n\nThe person who invited me was good in NLP and so we thought it would be great. After joining I found that out of four, 2 members have already gone completely silent. It was rest 3 of us. And guess what, they're from China (both). After that one has become competition expert without doing anything. After the competition, those users congratulated us when we posted our solution in the discussion. 😅\n\nYour findings clarify many things to me. I always wonder that how can `Joined 2/3/4/5/... days ago` surpass many top kaggler (the real heroes). I don't want to repeat much, but now I can tell most of them are from China/Beijing. 😯", "973668": "I hope private sharing doesn't break out like the corona epidemic :)", "973664": "Anything that has monetary gains will have scammers. I don't know how much of a priority Kaggle considers this to be but I won't be surprised if there is increasing amount of unfair practises in the whole system due to historic cases of cheaters getting away with it.\n\nBut certainly if this continues it will be a shame that Competitions (also) is not clean anymore and it will be curtains down for many.", "974008": "@divyansh22 These kind of inappropriate and hateful comments are not welcome here. I trust that you will reflect and do better.", "973997": "@divyansh22 Wow always think twice before posting this kind of s**t thanks", "973550": "😲😲😲 jonney has changed name to [jj-yytu](https://www.kaggle.com/swsuh2003) and he's in the team just before me ! \n\nPlus, he WAS EJECTED in [PANDA](https://www.kaggle.com/c/prostate-cancer-grade-assessment/discussion/169423#945549) and in [ALASKA2](url) (he was in silver in public and now inexist in LB I remember that because we were close).\n\nThat means Kaggle knows he cheated but kept his account just because he's a MASTER ?", "973486": "I don't understand who pays for Kaggle medals. I have never seen any benefit of having Kaggle medals.", "974004": "@divyansh22 Don't vent your anger on a specific country. Chinese competitors are not responsible for this small group of people. Your comment is totally racist. What has Kaggle to do COVID?\n\n(Kaggle discussions seem to be broken. I can't reply to comments)", "974656": "I think it's time to make at least ID and face verification as in EdX, at least. This will prevent such issues not 100%, but mostly.\n\nAlso it will help good Kagglers not to team up someone wrong who can cheat and crash the work. Happened to my team at least once (proven one), which led to loss of TOP3% silver and 2.5 month of work. ", "973783": "What the hell is this???????? They are totally undermining the efforts of many Kagglers. Kaggle should take strong action and ban all those users who got any single medal using such platform.", "974246": "Amazing investigative work! That's very bad for the Kaggle community and makes the ranking a little less meaningful, which is sad. ☹️\n\nThat being said, nothing beats the joy of learning something new and exploring new ideas, so I guess these that pay for medals won't get that. 😄\n\nFinally, if I may ask, do you know why you were contacted? Did it happen after you teamed up with the suspicious team-mate or was that just a coincidence?\n\nAgain, thanks for the post!", "974221": "The value of things is determined by how easy or hard is to get them. If it will be possible to simply buy Kaggle medals, their value will decrease. So we all, including Kaggle and competitors, are interested that cheaters are revealed and their medals removed.", "974115": "@inversion plenty of evidence here and also here https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/174635#973667 .\n\nedu.angtk.com took their website down as @group16 , I and others were challenging them last night. You see the main representative https://www.kaggle.com/lifeifei38 , who apparently is part of the staff for this medal farm, has an account a few months old with no activity. You can't let them get away with these medals in Wheat and Melanoma..........the lid is blown open on this group, and they have been doing this apparently for at least 2 years, but it's blatant, \"in your face\" now. I mean just look at the groups with user @yikououbao . Thats pure medal farming and yikououbao has taken real medals, and prevented many others from getting them. ", "973647": "I think the danger of this situation cannot be stressed enough: It threatens to undermine the community based on the open-source premise which makes it fair for all to compete, regardless of sociodemographic background. Open, and transparent collaboration strengthens the community by building trust in the methods and helping to push the cutting edge.\n\nCompanies / methods like this are toxic and damage the progress made by the wider open source community and the ideals they champion. Taking such large sums of money off people who may not know better is a low-grade business tactic - shame on them.\n\n $500 is a lot of money even for someone from the US/EU! For comparison, High-caliber ML courses vary from free to tens of dollars...", "973492": "The account you mention looks highly likely to be a medal farm that tricks its way onto high-scoring teams. I expect the photo was chosen to increase credibility and to influence people.", "974140": "As a side note, and for what it's worth, I feel like it would be perfectly fine for the private LB validation to take more time in general. This doesn't have to be an instantaneous switch; especially if the additional time helps to make sure that all the final placings are legit. As long as the time scales are clearly communicated, some delay might even help in keeping more people interested in post-submission deadline discussions and solution sharing.", "973730": "First of all, you did a great job finding this. \nSecond, Kaggle should take necessary steps against these users. It discourages beginners and disrupts the whole sharing and learning process among the community.", "973630": "In one of their blog post on ion switching competition: https://zhuanlan.zhihu.com/p/150559937 \n\nAn image caption as: \"top 15% can be reached using our early-stage solution\"\n\n\nIt's just blatant private sharing?", "973592": "This is becoming Kaggle CSI, indeed it is very hard to catch this things in a systematic way with concrete proof.\n\nAnyway, it is interesting that people think that Kaggle medals will give them jobs, just by having them, any serious company will make sure you know about things, a lot of companies don't even care about Kaggle.", "976179": "Jumped up 5 places, cheaters getting rid of?", "974310": "It seems the LB is being screwed up again....\n\nI see many novice in solo and teams who jumped lately in range 70th to 80th with very few submissions. \n\nFor instance [this one](https://www.kaggle.com/catting) just suscribed 10 days ago. ", "973509": "That's seriously Awful, I feel pity for those who ask for upvotes on LinkedIn after seeing this, those guys deserve way more respect than these people. ", "973493": "You did a great job finding this. I haven't participated in this competition, but I have been following what the community has been doing. I honestly was feared seeing how an open platform is being exploited.\n\nThe last 24 hours were a mess. Private sharing and the $1000 course.\n\nIt's great to see people like you and @brianfeeny bringing up some conclusive evidence. I hope the Kaggle team works on this issue, please don't let this happen in the future competitions.\n\nPeople work hard to get themselves in the top, it's disheartening to see their hardwork wash away by some malicious people.\n\nWhat the Kaggle team can do for the future:\n- Remove high-scoring kernels during the last days of the competition\n- Verify participants' accounts with real user data\n- Set up a robust anti-cheat system\n- Have a moderator who is quick to remove non-compliant submissions", "974883": "Thanks for @group16 & other Kagglers to dig this out. Definitely support for necessary actions to clean our community. Besides such cheating for medals, I'm also speechless about \"cheating for upvotes via folked notebooks\" activities in recent competitions. \n\nI'm thrilled to see so many GM's are in this competition and look forward to learning from their high-quality kernels/notebooks. E.g. Chris did a fantastic job and I learnt so much from the series of notebooks. But I feel uncomfortable to see people \"copy & paste\" others' original work, reproduce some \"high-score kernels\" with same method, input and logic, and ask for upvotes presumptuously. In the Wheat detection, I've seen such activity many times. Sometimes the post is not even related to the competition and the author would just scam everywhere and reply \"Great work! Check and upvote my EDA/notebook here!\"....OK, a notebook grandmaster/master/expert should be also a great proof of efforts, totally agree with that. But earning it in this way has no difference from antag.ai people actually. I know this is some \"self-marketing\" tricks, but getting a Stanford online certificate does not make real Stanford alumni.\n\nAfter going through the comments, what drove me a little bit more uncomfortable is the \"hateful\" stereotype towards a specific country. Now the issue becomes a counterpart of coronavirus in Kaggle. When the pandemic firstly starts, some people call it \"China virus\". But should we call it \"US virus\" now? What matters is the \"vaccine\", not the name for the virus...", "979044": "Dear fellow Sherlocks\n\nthose two accounts seem very much to belong to just one person:\nhttps://www.kaggle.com/jiuntianrobot\nhttps://www.kaggle.com/jiuntian\n\nThey follow each other & the robot account only handed in solutions on those days where jiuntian was handing in solutions.\nToo bad for our colleague: The robot chose the final submissions more wisely ( = higher rank ;))\n\n\nhttps://www.kaggle.com/tempusername1234 <-- looks also suspicious, but I cant link him yet to our magnificent 7 medal-buyers.\n\nhttps://www.kaggle.com/qwert217 <-- same", "975476": "Shocking that people are actually buying and selling kaggle positions! Although I guess that's not too surprising given how they do have value. Definitely agree with what others have said about Kaggle requiring ID verification like Coursera or EdX.", "975122": "very bad.. but all those have serious learning intent remain unmoved by such medal based temptations and they continue working hard ,take quidances from expert and keep moving..", "974661": "@group16 amazing job\nHow you found this? The difficulty to search and detect suspicious stuff from this part of the world is alphabet. It complicates the search. But I see this brilliant work and appreciate that. As more such job done, as better Kaggle will be for future participants. I hope. ", "974100": "Rather than delete cheaters, why not keep them in a public hall of shame. That way if they're discovered and verified as being cheaters, it can always be cross-referenced both now and in the future.", "974099": "Interesting @group16. Thanks for investigation and sharing.\nKaggle should step up to improve/keep its reputation. ", "974029": "I don't think it is possible for Kaggle team to check every single entry it will take forever to have the result of the competitions, I think it is best to verify novice account that gets a medal (ask them for account verification and a code check), in a future competition give the possibility for the Kaggle community(preferably expert Kaggler) to report a user for a code check, for example, they get more than 20 reports or something like that the reported team will need to perform a code and identity check. Just trying to suggest some ideas ", "973940": "Great work @group16 . This kind of cheating will destroy Kaggle if let continue. It also seems like more screening is needed before users can participate in competitions that award medals and prize money. ", "976771": "@group16, this is amazing. Well done. \n\nI think Kaggle needs a team of private investigators to tackle this fairly new crime and you should be hired right away.", "975483": "People taking shortcuts will never learn. ", "974014": "I think Kaggle needs to not allow novice account to join competitions until they become a Contributor. just to make it harder to cheat ", "973941": "As someone who spent a ridiculous amount of time to barely scrape through to get my sole silver medal and competition expert status, this kind of behavior makes me incredibly sad. Thank you @group16 for bringing this to light and exposing these cheaters.", "973675": "Its scary to be precise. I hope these accounts get blocked.. @group16 - Hopefully, the particular person in your team has not been involved in private sharing, else your complete 3 months effort would be futile.. Appreciate your efforts in bringing this up. ", "973639": "@daishu wrote this previous year.\n\n> There are some people who get solutions by joining different teams, and then improve their scores by private sharing. In China, we call this behavior \"PY transaction\", a typical group is \"宝典帮\". Now they are trying to pollute kaggle's competition environment, and they are sneaking into these teams:\n\nThose people might be joining multiple teams and collecting information from all of them just like before. Do you think your missing teammate can be in a circle like this? @group16", "973478": "yeah , so many novice accounts in gold medal range makes thing fishy", "975931": "Makes me wonder what Data Science interviews in China are like.", "974060": "@divyansh22 I feel sorry for you. After so many years of education, how do you react to things? Blame the whole country? Please don't embarrass yourself.", "973937": "gotta catch' em all!", "973932": "Thanks for you and your anonymous source's work. You've done the right thing. What happens in this competiton's public LB is CRAZING. Some people are getting more and more fearless that they now do not know what hiding means. A short and yet meanless way for money and fame drove them crazy.\n\nI wish kaggle can take swift action in this competition along with long-term policies to stop things from happening again. I also understand that concrete evidence is extremely hard to find in this kind of situation and every policy relating to the whole community is hard to make. So thanks for kaggle team's great work behind the scene.\n\nThe job needs to be done. Decaying and destroying is much more faster than growing and building. A snapshot of the current public LB can be a tragedy across the Internet. **What kaggle does next matters in many ways: for kaggle itself; for organizers; for this wonderful community; for all other honest and devoted competitors; and for other communities in the spirit of fair sharing.** The task could be heavy.\n\nBest wish for kaggle and kagglers!", "973545": "AFAIK the angtk.ai I knew was originally a data science community where people share general topics in data science (kinda like ods.ai) and help each other learn a few year back when I was still in high school. But then I got caught up in school work and didn't check their forum (and kaggle as well) for quite a while. Seems like they've evolved into this shady organization that profits (?) off of private sharing. What a pity. Or there may be two angtk.ai's and I was checking out the wrong one.", "974056": "I'm really really really sorry for my comment made on this post!!! I didn't hear what I was saying!! I really want to remove that comment but I am kinda unable to delete the discussion. There's a glitch with the discussions... I never intended to hurt anyone's feelings! I know I have hurt a lot of Kagglers feelings!! I'm really really sorry once again for my actions! No hard feelings on any specific community... I couldn't agree more that I sound really racist and I wish I could undo that! @kaggleteam, I really want to get things sorted, please look into the matter.🙏 Once again a deep regret and sorry for my actions!! I hope people forgive me, especially the ones I've offended!!🙏", "980592": "I'm new to Kaggle and I'm glad I came across this post. It's really interesting that someone would buy a rank. It's kind of like someone buying a college degree without ever attending. When it comes down to it, if someone gets a job partially based on a Kaggle user rating, they may not be able to keep it due to lack of ability.", "975542": "I always thought the purpose of this platform is to improve one's coding skill and try to make our best on difficult competitions... apparently not, i would never imagine such an industry could be possible...", "974142": "My god, the evidence just keeps mounting. Now we have a confession of a Kaggle master in the discussions. Basically anyone associated with angtk.com seems part of this ring. Either you are being paid, as a Master/GM, to take on newbies into your team, who can contribute nothing, so they can get medals..........you can claim ignorance......and that you were just doing it for advertising. There also appears to be code sharing/selling which is distributed through some private method, Kaggle accounts are created with no bio, no activity, they do not participate in discussions, comments, or anything. These accounts submit top scores.....they could be used to lower others, or to probe or to try to go for medals. Some of the accounts have never been used yet. All associated with angtk.com.", "974044": "@group16 Its sad that there are people using kaggle to sell medals and not to learn. At least there are still people using kaggle to learn and not to cheat.", "974011": "@stanleyjzheng Kaggle discussions are broken. I hope this gets fixed.", "973890": "That's ridiculous! It even doesn't make any sense since people learn NOTHING from these kind of sources, just ruin the other's competition experience", "973874": "this is pretty scary, thanks for sharing. Hope you can get a better teammate who wont ghost you", "973843": "kaggle can implement something like an user account should be old enough before joining any competitions.", "973794": "If Kaggle workers don't ban them now they will do it over and over again!", "973681": "Kaggle has to take a decision regarding this issue bt tracing the Profiles, it doesn't make sense that with only few numbers of submissions less than 50 submissions to reach first five.", "973679": "So, if I understand correctly, they use these fake accounts for testing submissions. They choose other (usually low-score submits for final score) and they sell high-scores submissions to the third party. \n\nYou raised a very important question, let's solve it together. Maybe Kaggle workers will be forced to rebuild the system of checking users and their submissions. To be honest, there is no system at all. It's time to build it.\n\nThere are two ways how to fight with that:\n1) Find out who is behind all this and punish them.\n2) Find out cheating solutions and ban them.\nI think the first one is much more effective. We should fight not with a cause but with a reason.\n", "973618": "It also works the other way around. \nSee if you have students who will pay for submissions... it gives the cheaters the advantage of improving their scores as they can use the information from the public lb of their students.\n\nThe students are actually farmers who rented a piece of land and must also share some of their crops with their landlords.", "973598": "Thanks for raising awareness on this! I did not know such a phenomenon existed in Kaggle or rather I didn't pay much attention to it until now. I think the most important thing is to learn as a community, we should come up with ways of detecting such misbehavior and penalize these users", "973523": "I remember something similar happened in the IEEE-CIS Fraud detection competition (almost a year ago) as well. ", "974789": "**Listen straight from the horse's mouth**\n\n> I've given some notebook and docs from past competitions as examples. And they pay me for this. At the same time, they asked me to add the names of some of the teams[ angtk.ai ]. pay for this. But I don't know what kind of ads they use these screenshots for. At the same time, advertisements from more than one source cannot be used as evidence.\n\nhttps://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/175201\n\n> I don't know where these screenshots came from. I've shared with you some of the cases that have ended the competition angtk.ai They also asked me to add the team name[ angtk.ai ]And pay for it. I didn't do anything else. Which screenshots came from, or from PS, can't be confirmed.\n\nhttps://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/175196\n\n> I sold them some code that had ended the game as a case study. They asked me to put it before the team name[ angtk.ai ]As an advertisement and pay for it! They used a screenshot of my team as an ad, but I didn't confirm the content carefully. I think it's really strange to say that I have private use with some screenshots of advertisements. In China, most advertisements exaggerate!\n\nhttps://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/175205\n\n", "974514": "Based on the results, Now we can say that Public LB is totally screwed up but **\"Hard work will never fail\".** I like that the Kaggle team has worked on and yah still working on removing violaters.", "974081": "@group16 thanks for the breakdown and here and in https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/174635#973667 . This is no small problem. In this competition alone it seems there may be dozens of people farming medals, many at the top of the LB.", "973989": "Oh my god, I double checked if you were Chinese.😹\nThis is really disgusting to buy a Kaggle medal! It made so many people's hard work end up with nothing.", "973945": "Vopani\"s words are wise: \n\n> Anything that has monetary gains will have scammers ", "974070": "@yuanlin08 I have nothing to do with education, it just the perfect example for when u talk out of anger, he did apologize for it no need to be that hard on him.", "978980": "nice very!!!!!", "979201": "nice......", "978911": "@group16 , Great !!!", "987357": "I'm new to kaggle but buying ranks for fake knowledge depiction would never ever work in life.", "986630": "I'm new to Kaggle and I'm glad I came across this post good job ", "985889": "Kaggle Please take some actions on these type of fraud practices.\n", "981997": "maybe Kaggle admins can use anomaly detection tech based on user's notebook logs. Very few cases can get a high mark in first-run unless someone copied the code or model. ", "979068": "I want to see write-ups of solution from teams @siim2020_stacking, @sunlight and @zhaigongzuo2", "975895": "It is so awful, kaggle maybe improve the competition enviornment.", "974367": "Thank you for heads up.\nReally like the idea of metadata graphing too.\nAfter comp deadline, before any results published or winners announced and remove all cheats.", "974266": "I guess this website is also used to share content? https://www.angtk.com/", "974212": "Can I assume this type of fraud activity is one of the kind of reasons why genuine kaggle grandmasters or genuine competitors leaving this place to compete?", "974085": "I have a feeling that we will wait a lot more than usual for the final private lb and medals :) ", "974020": "Thanks for this information :) This is quite a revelation for someone who started a week ago. Hope the community finds a way to circumvent this. ", "973947": "Shame on them", "973568": "You should work at kaggle's compliance department :) \nOne might think that it would be easy for Google/Kaggle with its resources to find cheaters, especially if someone flag with a name where a solution was bought if it is as you say. And why signal the misstake in the name, all strange.\nHave you reported the teams / profiles / notebooks to Kaggle? then they can certainly do their thorough investigation.", "981181": "", "979451": "", "976904": "", "976901": "", "975918": "", "974403": "", "974009": "", "973761": "", "973678": "", "973677": "", "973578": "", "976704": "Cool. Thank you." } }