{
  "id": 19664,
  "title": "Randomizing data",
  "url": "/competitions/expedia-personalized-sort/discussion/19664",
  "author_name": "",
  "post_date": "2016-03-20T16:17:52.113Z",
  "votes": null,
  "comment_count": 2,
  "views": 656,
  "content": "<p>Hello, </p>\n\n<p>I saw that there are several posts where people discussed about position\nbias and randomization of the data, but still not quite sure if the data can be\nassumed if it is randomized.</p>\n\n<p>In the data introduction page, Expedia specified, &quot;the data contain impressions where the hotels were randomly sorted, to avoid the position bias of the existing algorithm&quot;\nand I also saw the post in this forum that you included randomly sorted hotels to make less biased data. </p>\n\n<p>Based on the discussion, I understand that Expedia puts an effort to make the\ndata randomized. Does I understand correctly? If that is the case, do you think \nthat I can assume that the data was randomly drawn from the population? That is, can I assume that there is no systematic bias in the data?</p>\n\n<p>Thanks you so much for your attention to this post and hope to hear from you.</p>\n\n<p>Thanks!</p>",
  "messages": [
    {
      "id": "112370",
      "postDate": "03/20/2016 16:17:52",
      "content": "<p>Hello, </p>\n\n<p>I saw that there are several posts where people discussed about position\nbias and randomization of the data, but still not quite sure if the data can be\nassumed if it is randomized.</p>\n\n<p>In the data introduction page, Expedia specified, &quot;the data contain impressions where the hotels were randomly sorted, to avoid the position bias of the existing algorithm&quot;\nand I also saw the post in this forum that you included randomly sorted hotels to make less biased data. </p>\n\n<p>Based on the discussion, I understand that Expedia puts an effort to make the\ndata randomized. Does I understand correctly? If that is the case, do you think \nthat I can assume that the data was randomly drawn from the population? That is, can I assume that there is no systematic bias in the data?</p>\n\n<p>Thanks you so much for your attention to this post and hope to hear from you.</p>\n\n<p>Thanks!</p>",
      "rawMarkdown": "Hello, \r\n\r\nI saw that there are several posts where people discussed about position\r\nbias and randomization of the data, but still not quite sure if the data can be\r\nassumed if it is randomized.\r\n\r\nIn the data introduction page, Expedia specified, \"the data contain impressions where the hotels were randomly sorted, to avoid the position bias of the existing algorithm\"\r\nand I also saw the post in this forum that you included randomly sorted hotels to make less biased data. \r\n\r\nBased on the discussion, I understand that Expedia puts an effort to make the\r\ndata randomized. Does I understand correctly? If that is the case, do you think \r\nthat I can assume that the data was randomly drawn from the population? That is, can I assume that there is no systematic bias in the data?\r\n\r\nThanks you so much for your attention to this post and hope to hear from you.\r\n\r\nThanks!",
      "votes": null
    },
    {
      "id": "112440",
      "postDate": "03/21/2016 09:31:37",
      "content": "<p>Yes, you can assume that data was randomly drawn from the population. However, some impressions were incorrectly labelled as random, so you probably need to remove these outliers.</p>\n\n<p>Thanks,\nAdam</p>",
      "rawMarkdown": "Yes, you can assume that data was randomly drawn from the population. However, some impressions were incorrectly labelled as random, so you probably need to remove these outliers.\r\n\r\nThanks,\r\nAdam",
      "votes": null
    },
    {
      "id": "112667",
      "postDate": "03/22/2016 18:16:25",
      "content": "<p>Thank you so much, Adam!</p>\n\n<p>Best regards,\nEunhee Emily Ko</p>",
      "rawMarkdown": "Thank you so much, Adam!\r\n\r\nBest regards,\r\nEunhee Emily Ko",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 112440,
      "author_name": "adamwoz",
      "author_url": "",
      "post_date": "03/21/2016 09:31:37",
      "content": "<p>Yes, you can assume that data was randomly drawn from the population. However, some impressions were incorrectly labelled as random, so you probably need to remove these outliers.</p>\n\n<p>Thanks,\nAdam</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 112667,
      "author_name": "eunheeko",
      "author_url": "",
      "post_date": "03/22/2016 18:16:25",
      "content": "<p>Thank you so much, Adam!</p>\n\n<p>Best regards,\nEunhee Emily Ko</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "112370": "Hello, \r\n\r\nI saw that there are several posts where people discussed about position\r\nbias and randomization of the data, but still not quite sure if the data can be\r\nassumed if it is randomized.\r\n\r\nIn the data introduction page, Expedia specified, \"the data contain impressions where the hotels were randomly sorted, to avoid the position bias of the existing algorithm\"\r\nand I also saw the post in this forum that you included randomly sorted hotels to make less biased data. \r\n\r\nBased on the discussion, I understand that Expedia puts an effort to make the\r\ndata randomized. Does I understand correctly? If that is the case, do you think \r\nthat I can assume that the data was randomly drawn from the population? That is, can I assume that there is no systematic bias in the data?\r\n\r\nThanks you so much for your attention to this post and hope to hear from you.\r\n\r\nThanks!",
    "112440": "Yes, you can assume that data was randomly drawn from the population. However, some impressions were incorrectly labelled as random, so you probably need to remove these outliers.\r\n\r\nThanks,\r\nAdam",
    "112667": "Thank you so much, Adam!\r\n\r\nBest regards,\r\nEunhee Emily Ko"
  },
  "source": "meta"
}