{
  "id": 5205,
  "title": "Public Info about Pairs",
  "url": "/competitions/cause-effect-pairs/discussion/5205",
  "author_name": "",
  "post_date": "2013-07-26T16:40:03.593Z",
  "votes": null,
  "comment_count": 2,
  "views": 1090,
  "content": "<p>Hello everybody,</p>\n<p>Someone has any idea to improve the model using the public info about the pairs? Or the target details</p>\n<p>&nbsp;</p>\n<p>Thanks in advance</p>\n<p></p>",
  "messages": [
    {
      "id": "27693",
      "postDate": "07/26/2013 16:40:03",
      "content": "<p>Hello everybody,</p>\n<p>Someone has any idea to improve the model using the public info about the pairs? Or the target details</p>\n<p>&nbsp;</p>\n<p>Thanks in advance</p>\n<p></p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "27717",
      "postDate": "07/27/2013 10:54:32",
      "content": "<p>The simpliest idea is to add these data as a new columns to training and validation datasets i.e. using DictVectorizer in python.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "28164",
      "postDate": "08/04/2013 00:33:42",
      "content": "<p>A model selection depending on the combination of these types could also be of advantage.</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 27717,
      "author_name": "vnefedov",
      "author_url": "",
      "post_date": "07/27/2013 10:54:32",
      "content": "<p>The simpliest idea is to add these data as a new columns to training and validation datasets i.e. using DictVectorizer in python.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 28164,
      "author_name": "analyticbastard",
      "author_url": "",
      "post_date": "08/04/2013 00:33:42",
      "content": "<p>A model selection depending on the combination of these types could also be of advantage.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "27693": "",
    "27717": "",
    "28164": ""
  },
  "source": "meta"
}