{
  "id": 21138,
  "title": "A question about xgboost?",
  "url": "/competitions/expedia-hotel-recommendations/discussion/21138",
  "author_name": "dot277",
  "post_date": "2016-05-22T13:57:57.197000",
  "votes": 0,
  "comment_count": 0,
  "views": 391,
  "content": "<p>How does xgboost works for classifcation? \nIn <a href=\"http://xgboost.readthedocs.io/en/latest/model.html\">http://xgboost.readthedocs.io/en/latest/model.html</a></p>\n\n<p>it is presented as a tree ensemble, but than it says it can support weighted logistic regression loss. From what i know about trees, they work with the information gain as an optimization problem.</p>\n\n<p>I would love a clarification.</p>",
  "messages": [
    {
      "id": 120994,
      "postDate": "2016-05-22T13:57:57.197Z",
      "content": "<p>How does xgboost works for classifcation? \nIn <a href=\"http://xgboost.readthedocs.io/en/latest/model.html\">http://xgboost.readthedocs.io/en/latest/model.html</a></p>\n\n<p>it is presented as a tree ensemble, but than it says it can support weighted logistic regression loss. From what i know about trees, they work with the information gain as an optimization problem.</p>\n\n<p>I would love a clarification.</p>",
      "rawMarkdown": "How does xgboost works for classifcation? \r\nIn http://xgboost.readthedocs.io/en/latest/model.html\r\n\r\nit is presented as a tree ensemble, but than it says it can support weighted logistic regression loss. From what i know about trees, they work with the information gain as an optimization problem.\r\n\r\nI would love a clarification."
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "120994": "How does xgboost works for classifcation? \r\nIn http://xgboost.readthedocs.io/en/latest/model.html\r\n\r\nit is presented as a tree ensemble, but than it says it can support weighted logistic regression loss. From what i know about trees, they work with the information gain as an optimization problem.\r\n\r\nI would love a clarification."
  }
}