{
  "id": 2159,
  "title": " the leaderboard-text.json file ",
  "url": "/competitions/cprod1/discussion/2159",
  "author_name": "",
  "post_date": "2012-07-10T10:49:53.093Z",
  "votes": null,
  "comment_count": 1,
  "views": 1499,
  "content": "<p>Excuse me !,the interpretation of the&nbsp; leaderboard-text.json is &quot;This .json file contains the text items that contestants must disambiguate to determine their leaderboard score.&nbsp;&quot;&nbsp;.&nbsp;&nbsp;But&nbsp;the content of the&nbsp;leaderboard-test.json&nbsp;&nbsp;is also text-items,&nbsp; so what\r\n the distinction of the leaderboard-text.json and the training-non-annotated-text.json</p>",
  "messages": [
    {
      "id": "12258",
      "postDate": "07/10/2012 10:49:53",
      "content": "<p>Excuse me !,the interpretation of the&nbsp; leaderboard-text.json is &quot;This .json file contains the text items that contestants must disambiguate to determine their leaderboard score.&nbsp;&quot;&nbsp;.&nbsp;&nbsp;But&nbsp;the content of the&nbsp;leaderboard-test.json&nbsp;&nbsp;is also text-items,&nbsp; so what\r\n the distinction of the leaderboard-text.json and the training-non-annotated-text.json</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "12261",
      "postDate": "07/10/2012 12:42:37",
      "content": "<p>The relationship between the different .json text(item) files is as follows: All text_item files were drawn from the same large population of underlying web content. We first attempted to remove any duplicate content and then randomly assigned each text\r\n item to a single file.</p>\r\n<p>Cheers.</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 12261,
      "author_name": "cprod1",
      "author_url": "",
      "post_date": "07/10/2012 12:42:37",
      "content": "<p>The relationship between the different .json text(item) files is as follows: All text_item files were drawn from the same large population of underlying web content. We first attempted to remove any duplicate content and then randomly assigned each text\r\n item to a single file.</p>\r\n<p>Cheers.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "12258": "",
    "12261": ""
  },
  "source": "meta"
}