{
  "id": 321660,
  "title": "How many recall methods are available?",
  "url": "/competitions/h-and-m-personalized-fashion-recommendations/discussion/321660",
  "author_name": "",
  "post_date": "2022-04-28T03:34:57.796000",
  "votes": 27,
  "comment_count": 12,
  "views": 0,
  "content": "<p>Mine: </p>\n<ol>\n<li>popular articles in recently; </li>\n<li>the articles which the customer bought in history;</li>\n<li>itemcf;</li>\n<li>usercf;</li>\n</ol>",
  "messages": [
    {
      "id": 1770152,
      "postDate": "2022-04-28T03:34:57.797Z",
      "content": "<p>Mine: </p>\n<ol>\n<li>popular articles in recently; </li>\n<li>the articles which the customer bought in history;</li>\n<li>itemcf;</li>\n<li>usercf;</li>\n</ol>",
      "rawMarkdown": "Mine: \n1. popular articles in recently; \n2. the articles which the customer bought in history;\n3. itemcf;\n4. usercf;",
      "votes": 26
    },
    {
      "id": 1771145,
      "postDate": "2022-04-29T00:19:58.337Z",
      "content": "<p>update: <br>\nprice_popular: calculate the mean price of user, and get the most popular item in every price bin, <br>\nfor every user, find its bin and recommend most popular items in this bin.</p>",
      "rawMarkdown": "update: \nprice_popular: calculate the mean price of user, and get the most popular item in every price bin, \nfor every user, find its bin and recommend most popular items in this bin.",
      "votes": 13,
      "replies": [
        {
          "id": 1775342,
          "postDate": "2022-05-02T22:56:54.517Z",
          "content": "<p>It looks helped</p>",
          "rawMarkdown": "It looks helped"
        }
      ]
    },
    {
      "id": 1770203,
      "postDate": "2022-04-28T04:51:36.953Z",
      "content": "<p>Same 1,2,3 with yours.<br>\nAn additional user tag recall: popular items under user's attributes (age, gender and so on)<br>\nI didn't use usercf and will try it later<br>\nthank you for sharing!</p>",
      "rawMarkdown": "Same 1,2,3 with yours.\nAn additional user tag recall: popular items under user's attributes (age, gender and so on)\nI didn't use usercf and will try it later\nthank you for sharing!",
      "votes": 6,
      "replies": [
        {
          "id": 1774838,
          "postDate": "2022-05-02T13:41:16.883Z",
          "content": "<p>thanks for your sharing, by the way, how can you get gender and \"so on\" attributes of user, I can't find them in user profile…</p>",
          "rawMarkdown": "thanks for your sharing, by the way, how can you get gender and \"so on\" attributes of user, I can't find them in user profile..."
        },
        {
          "id": 1775000,
          "postDate": "2022-05-02T16:25:16.687Z",
          "content": "<p>according to what he/she bought in history</p>",
          "rawMarkdown": "according to what he/she bought in history",
          "votes": 1
        },
        {
          "id": 1775246,
          "postDate": "2022-05-02T20:36:15.120Z",
          "content": "<p>Is the usercf worthwhile? Intuitionally I felt we would get very small overlaps to infer anything useful.</p>",
          "rawMarkdown": "Is the usercf worthwhile? Intuitionally I felt we would get very small overlaps to infer anything useful."
        }
      ]
    },
    {
      "id": 1771642,
      "postDate": "2022-04-29T12:42:34.437Z",
      "content": "<p>I think it's necessary to categorize the buyer's age, gender, etc., and factor that into the strategy.</p>\n<p>Thank you for sharing!</p>",
      "rawMarkdown": "I think it's necessary to categorize the buyer's age, gender, etc., and factor that into the strategy.\n\nThank you for sharing!",
      "votes": 3
    },
    {
      "id": 1774863,
      "postDate": "2022-05-02T14:18:00.083Z",
      "content": "<p>Hi, what does itemcf and usercf mean? Can someone explain? </p>",
      "rawMarkdown": "Hi, what does itemcf and usercf mean? Can someone explain? ",
      "votes": 1,
      "replies": [
        {
          "id": 1775163,
          "postDate": "2022-05-02T18:40:43.107Z",
          "content": "<p>I think cf stands for collaborative filtering in this case.</p>",
          "rawMarkdown": "I think cf stands for collaborative filtering in this case.",
          "votes": 1
        }
      ]
    },
    {
      "id": 1775497,
      "postDate": "2022-05-03T05:25:20.107Z",
      "content": "<p>May I ask how many products did you recall for each user?</p>",
      "rawMarkdown": "May I ask how many products did you recall for each user?",
      "votes": 1
    },
    {
      "id": 1771231,
      "postDate": "2022-04-29T02:44:21.660Z",
      "content": "<p>Thanks for your sharing! And I'm curious about what your recall score is based on your recall methods, in my recall strategy, the recall score is about 0.11 for 100 candidate articles, and I don't know if it's normal.</p>",
      "rawMarkdown": "Thanks for your sharing! And I'm curious about what your recall score is based on your recall methods, in my recall strategy, the recall score is about 0.11 for 100 candidate articles, and I don't know if it's normal."
    },
    {
      "id": 1774868,
      "postDate": "2022-05-02T14:23:13.250Z",
      "rawMarkdown": "",
      "isDeleted": true
    }
  ],
  "comments": [
    {
      "id": 1771145,
      "author_name": "",
      "author_url": "",
      "post_date": "2022-04-29T00:19:58.337000",
      "content": "<p>update: <br>\nprice_popular: calculate the mean price of user, and get the most popular item in every price bin, <br>\nfor every user, find its bin and recommend most popular items in this bin.</p>",
      "votes": 13,
      "replies": [
        {
          "id": 1775342,
          "author_name": "lemon",
          "author_url": "",
          "post_date": "2022-05-02T22:56:54.517000",
          "content": "<p>It looks helped</p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 1770203,
      "author_name": "sirius",
      "author_url": "",
      "post_date": "2022-04-28T04:51:36.953000",
      "content": "<p>Same 1,2,3 with yours.<br>\nAn additional user tag recall: popular items under user's attributes (age, gender and so on)<br>\nI didn't use usercf and will try it later<br>\nthank you for sharing!</p>",
      "votes": 6,
      "replies": [
        {
          "id": 1774838,
          "author_name": "biubiuG",
          "author_url": "",
          "post_date": "2022-05-02T13:41:16.883000",
          "content": "<p>thanks for your sharing, by the way, how can you get gender and \"so on\" attributes of user, I can't find them in user profile…</p>",
          "votes": 0,
          "replies": []
        },
        {
          "id": 1775000,
          "author_name": "sirius",
          "author_url": "",
          "post_date": "2022-05-02T16:25:16.687000",
          "content": "<p>according to what he/she bought in history</p>",
          "votes": 1,
          "replies": []
        },
        {
          "id": 1775246,
          "author_name": "Mayukh Bhattacharyya",
          "author_url": "",
          "post_date": "2022-05-02T20:36:15.120000",
          "content": "<p>Is the usercf worthwhile? Intuitionally I felt we would get very small overlaps to infer anything useful.</p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 1771642,
      "author_name": "",
      "author_url": "",
      "post_date": "2022-04-29T12:42:34.437000",
      "content": "<p>I think it's necessary to categorize the buyer's age, gender, etc., and factor that into the strategy.</p>\n<p>Thank you for sharing!</p>",
      "votes": 3,
      "replies": []
    },
    {
      "id": 1774863,
      "author_name": "Cai Zhuo",
      "author_url": "",
      "post_date": "2022-05-02T14:18:00.083000",
      "content": "<p>Hi, what does itemcf and usercf mean? Can someone explain? </p>",
      "votes": 1,
      "replies": [
        {
          "id": 1775163,
          "author_name": "Igor Kuivjogi Fernandes",
          "author_url": "",
          "post_date": "2022-05-02T18:40:43.107000",
          "content": "<p>I think cf stands for collaborative filtering in this case.</p>",
          "votes": 1,
          "replies": []
        }
      ]
    },
    {
      "id": 1775497,
      "author_name": "yuanzhe zhou",
      "author_url": "",
      "post_date": "2022-05-03T05:25:20.107000",
      "content": "<p>May I ask how many products did you recall for each user?</p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 1771231,
      "author_name": "Rock",
      "author_url": "",
      "post_date": "2022-04-29T02:44:21.660000",
      "content": "<p>Thanks for your sharing! And I'm curious about what your recall score is based on your recall methods, in my recall strategy, the recall score is about 0.11 for 100 candidate articles, and I don't know if it's normal.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1774868,
      "author_name": "",
      "author_url": "",
      "post_date": "2022-05-02T14:23:13.250000",
      "content": "",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1770152": "Mine: \n1. popular articles in recently; \n2. the articles which the customer bought in history;\n3. itemcf;\n4. usercf;",
    "1771145": "update: \nprice_popular: calculate the mean price of user, and get the most popular item in every price bin, \nfor every user, find its bin and recommend most popular items in this bin.",
    "1770203": "Same 1,2,3 with yours.\nAn additional user tag recall: popular items under user's attributes (age, gender and so on)\nI didn't use usercf and will try it later\nthank you for sharing!",
    "1771642": "I think it's necessary to categorize the buyer's age, gender, etc., and factor that into the strategy.\n\nThank you for sharing!",
    "1774863": "Hi, what does itemcf and usercf mean? Can someone explain? ",
    "1775497": "May I ask how many products did you recall for each user?",
    "1771231": "Thanks for your sharing! And I'm curious about what your recall score is based on your recall methods, in my recall strategy, the recall score is about 0.11 for 100 candidate articles, and I don't know if it's normal.",
    "1774868": ""
  }
}