{
  "id": 6122,
  "title": "Dealing with missing data",
  "url": "/competitions/expedia-personalized-sort/discussion/6122",
  "author_name": "",
  "post_date": "2013-10-25T20:52:27.687Z",
  "votes": null,
  "comment_count": 2,
  "views": 1391,
  "content": "<p>I could see that there are lots of missing values in the dataset. I wanted to discuss with you guys what sort of techniques you guys are using to handle this issue.</p>",
  "messages": [
    {
      "id": "32696",
      "postDate": "10/25/2013 20:52:27",
      "content": "<p>I could see that there are lots of missing values in the dataset. I wanted to discuss with you guys what sort of techniques you guys are using to handle this issue.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "32792",
      "postDate": "10/28/2013 19:22:02",
      "content": "<h2>I am simply replacing missing values with the mean value. For categorical data, I put missing values into their own category.</h2>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "32801",
      "postDate": "10/28/2013 22:18:17",
      "content": "<p>I recommend using median instead of mean. More robust to outliers. You can additionally create another categorical column to mark NAs. And some methods (like gbm) deals with NA just fine.</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 32792,
      "author_name": "looschen",
      "author_url": "",
      "post_date": "10/28/2013 19:22:02",
      "content": "<h2>I am simply replacing missing values with the mean value. For categorical data, I put missing values into their own category.</h2>",
      "votes": null,
      "replies": []
    },
    {
      "id": 32801,
      "author_name": "leustagos",
      "author_url": "",
      "post_date": "10/28/2013 22:18:17",
      "content": "<p>I recommend using median instead of mean. More robust to outliers. You can additionally create another categorical column to mark NAs. And some methods (like gbm) deals with NA just fine.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "32696": "",
    "32792": "",
    "32801": ""
  },
  "source": "meta"
}