{
  "id": 11098,
  "title": "A question about the type of Feature",
  "url": "/competitions/avazu-ctr-prediction/discussion/11098",
  "author_name": "",
  "post_date": "2014-12-01T01:28:04.627Z",
  "votes": null,
  "comment_count": 3,
  "views": 1774,
  "content": "<p>Hi, All.</p>\n<p>I'm a new kaggler, and I'm confused about the type of anonymous feature C14, of which values vary from hundreds to 20k. It looks like numerical, could anyone tell me whether it's numerical or categorical? If it's categorical, is it ordinal or not?</p>\n<p>I think it's important to my model.</p>\n<p>Thanks in advance!</p>",
  "messages": [
    {
      "id": "59299",
      "postDate": "12/01/2014 01:28:04",
      "content": "<p>Hi, All.</p>\n<p>I'm a new kaggler, and I'm confused about the type of anonymous feature C14, of which values vary from hundreds to 20k. It looks like numerical, could anyone tell me whether it's numerical or categorical? If it's categorical, is it ordinal or not?</p>\n<p>I think it's important to my model.</p>\n<p>Thanks in advance!</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "59312",
      "postDate": "12/01/2014 08:14:08",
      "content": "<p>From the <a href=\"https://www.kaggle.com/c/avazu-ctr-prediction/data\">data description page</a>:</p>\n<p style=\"padding-left: 30px\">C14-C21 -- anonymized categorical variables</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "59323",
      "postDate": "12/01/2014 10:43:10",
      "content": "<p>@Matt</p>\n<p>Thank you!</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "73068",
      "postDate": "04/22/2015 02:20:17",
      "content": "<p>@Matt</p>\n<p>how to deal with the&nbsp;anonymized&nbsp;variables? I mean, we can't know the meaning of&nbsp;anonymized variables, does it influence the result?</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 59312,
      "author_name": "mattvonrohr",
      "author_url": "",
      "post_date": "12/01/2014 08:14:08",
      "content": "<p>From the <a href=\"https://www.kaggle.com/c/avazu-ctr-prediction/data\">data description page</a>:</p>\n<p style=\"padding-left: 30px\">C14-C21 -- anonymized categorical variables</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 59323,
      "author_name": "linhaoye",
      "author_url": "",
      "post_date": "12/01/2014 10:43:10",
      "content": "<p>@Matt</p>\n<p>Thank you!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 73068,
      "author_name": "kaka10",
      "author_url": "",
      "post_date": "04/22/2015 02:20:17",
      "content": "<p>@Matt</p>\n<p>how to deal with the&nbsp;anonymized&nbsp;variables? I mean, we can't know the meaning of&nbsp;anonymized variables, does it influence the result?</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "59299": "",
    "59312": "",
    "59323": "",
    "73068": ""
  },
  "source": "meta"
}