I’m new to Kaggle and a bit confused about plagiarism. If I use someone else’s shared code in parts of my analysis, does that count as plagiarism?
\nAlso, I’m curious why people share their code during competitions. I don’t quite understand why someone would share their best work while competing. Could anyone explain?
", "messages": [ { "id": "3006253", "postDate": "10/03/2024 23:45:00", "content": "I’m new to Kaggle and a bit confused about plagiarism. If I use someone else’s shared code in parts of my analysis, does that count as plagiarism?
\nAlso, I’m curious why people share their code during competitions. I don’t quite understand why someone would share their best work while competing. Could anyone explain?
", "rawMarkdown": "I’m new to Kaggle and a bit confused about plagiarism. If I use someone else’s shared code in parts of my analysis, does that count as plagiarism?\n\nAlso, I’m curious why people share their code during competitions. I don’t quite understand why someone would share their best work while competing. Could anyone explain?", "votes": null }, { "id": "3006422", "postDate": "10/04/2024 06:07:35", "content": "Plagiarism: no, because all published notebooks in the code section are automatically open-source.
\nWhy: for medals and fun, 'cause people love sharing knowledge. That published work is rarely gold-range quality. Usually, it's either not 'their best work' or that it is, but 'their best work' is far from the final gold range, so why not share and get a notebook gold medal instead? You should understand that a lot of people care only for gold medals, so they can even share silver-range work without any reservation if they perceive that this won't hurt their chance of winning a gold medal (either because they don't think they have a chance at gold, or because what they share is basic enough that any strong competitor would already be familiar with it).
@rouzbehzarin as a safety measure, you should always cite the source of your copied content always. I usually create a separate section named References in my kernels highlighting all referred content. I usually never copy-paste anything directly as well, to avoid such issues
", "rawMarkdown": "rouzbehzarin as a safety measure, you should always cite the source of your copied content always. I usually create a separate section named **References** in my kernels highlighting all referred content. I usually never copy-paste anything directly as well, to avoid such issues", "votes": null }, { "id": "3006748", "postDate": "10/04/2024 14:11:18", "content": "Thank you for your thorough and excellent explanation!
\nI have another question. I'm a beginner on Kaggle, but not in my data work. I want to work with a team that has more experience than me so I can learn from them. How can I find a team like that?
Thank you so much! I'll get right on it!
\nI have another question—how can I find someone more experienced than me who would be interested in having me join their team?
I always work alone, so idk; maybe on Discord?
\nIf you are good, I recommend starting alone. You can easily find teammates if you have a strong solution on the LB.
\nIf you want to join a strong team, you need to prove your worth first…on Kaggle. Most won't care about your 'previous experience'; rather, if you have strong past experience, you should be able to get a good LB result by yourself. This is the fastest and most solid way to find a strong team.
You need to ask a lot of people about teaming up and someone will surely take you in their team. Also, prepare yourself for a lone effort if your request to team up does not go through @rouzbehzarin as a backup option.
", "rawMarkdown": "You need to ask a lot of people about teaming up and someone will surely take you in their team. Also, prepare yourself for a lone effort if your request to team up does not go through @rouzbehzarin as a backup option.", "votes": null }, { "id": "3006813", "postDate": "10/04/2024 15:14:26", "content": "Thank you! That makes perfect sense to me!
", "rawMarkdown": "Thank you! That makes perfect sense to me!", "votes": null }, { "id": "3006814", "postDate": "10/04/2024 15:15:14", "content": "Thank you, Ravi! I truly appreciate your response!
", "rawMarkdown": "Thank you, Ravi! I truly appreciate your response!", "votes": null } ], "comments": [ { "id": 3006422, "author_name": "shlomoron", "author_url": "", "post_date": "10/04/2024 06:07:35", "content": "Plagiarism: no, because all published notebooks in the code section are automatically open-source.
\nWhy: for medals and fun, 'cause people love sharing knowledge. That published work is rarely gold-range quality. Usually, it's either not 'their best work' or that it is, but 'their best work' is far from the final gold range, so why not share and get a notebook gold medal instead? You should understand that a lot of people care only for gold medals, so they can even share silver-range work without any reservation if they perceive that this won't hurt their chance of winning a gold medal (either because they don't think they have a chance at gold, or because what they share is basic enough that any strong competitor would already be familiar with it).
Thank you for your thorough and excellent explanation!
\nI have another question. I'm a beginner on Kaggle, but not in my data work. I want to work with a team that has more experience than me so I can learn from them. How can I find a team like that?
I always work alone, so idk; maybe on Discord?
\nIf you are good, I recommend starting alone. You can easily find teammates if you have a strong solution on the LB.
\nIf you want to join a strong team, you need to prove your worth first…on Kaggle. Most won't care about your 'previous experience'; rather, if you have strong past experience, you should be able to get a good LB result by yourself. This is the fastest and most solid way to find a strong team.
Thank you! That makes perfect sense to me!
", "votes": null, "replies": [] } ] } ] } ] }, { "id": 3006596, "author_name": "ravi20076", "author_url": "", "post_date": "10/04/2024 11:14:38", "content": "@rouzbehzarin as a safety measure, you should always cite the source of your copied content always. I usually create a separate section named References in my kernels highlighting all referred content. I usually never copy-paste anything directly as well, to avoid such issues
", "votes": null, "replies": [ { "id": 3006757, "author_name": "rouzbehzarin", "author_url": "", "post_date": "10/04/2024 14:23:36", "content": "Thank you so much! I'll get right on it!
\nI have another question—how can I find someone more experienced than me who would be interested in having me join their team?
You need to ask a lot of people about teaming up and someone will surely take you in their team. Also, prepare yourself for a lone effort if your request to team up does not go through @rouzbehzarin as a backup option.
", "votes": null, "replies": [ { "id": 3006814, "author_name": "rouzbehzarin", "author_url": "", "post_date": "10/04/2024 15:15:14", "content": "Thank you, Ravi! I truly appreciate your response!
", "votes": null, "replies": [] } ] } ] } ] } ], "raw_markdown_by_id": { "3006253": "I’m new to Kaggle and a bit confused about plagiarism. If I use someone else’s shared code in parts of my analysis, does that count as plagiarism?\n\nAlso, I’m curious why people share their code during competitions. I don’t quite understand why someone would share their best work while competing. Could anyone explain?", "3006422": "Plagiarism: no, because all published notebooks in the code section are automatically open-source. \nWhy: for medals and fun, 'cause people love sharing knowledge. That published work is rarely gold-range quality. Usually, it's either not 'their best work' or that it is, but 'their best work' is far from the final gold range, so why not share and get a notebook gold medal instead? You should understand that a lot of people care only for gold medals, so they can even share silver-range work without any reservation if they perceive that this won't hurt their chance of winning a gold medal (either because they don't think they have a chance at gold, or because what they share is basic enough that any strong competitor would already be familiar with it).", "3006596": "rouzbehzarin as a safety measure, you should always cite the source of your copied content always. I usually create a separate section named **References** in my kernels highlighting all referred content. I usually never copy-paste anything directly as well, to avoid such issues", "3006748": "Thank you for your thorough and excellent explanation!\nI have another question. I'm a beginner on Kaggle, but not in my data work. I want to work with a team that has more experience than me so I can learn from them. How can I find a team like that?", "3006757": "Thank you so much! I'll get right on it!\nI have another question—how can I find someone more experienced than me who would be interested in having me join their team?", "3006803": "I always work alone, so idk; maybe on Discord?\nIf you are good, I recommend starting alone. You can easily find teammates if you have a strong solution on the LB.\nIf you want to join a strong team, you need to prove your worth first...on Kaggle. Most won't care about your 'previous experience'; rather, if you have strong past experience, you should be able to get a good LB result by yourself. This is the fastest and most solid way to find a strong team.", "3006809": "You need to ask a lot of people about teaming up and someone will surely take you in their team. Also, prepare yourself for a lone effort if your request to team up does not go through @rouzbehzarin as a backup option.", "3006813": "Thank you! That makes perfect sense to me!", "3006814": "Thank you, Ravi! I truly appreciate your response!" }, "source": "meta" }