{
  "id": 242443,
  "title": "Columns labeling ",
  "url": "/competitions/gct-apac-d-and-a-race/discussion/242443",
  "author_name": "",
  "post_date": "2021-05-29T02:24:48.870849Z",
  "votes": 1,
  "comment_count": 1,
  "views": 0,
  "content": "<p>if it is 'anonymized, binary classification dataset ' what is the point of column labeling in the data set? </p>\n<p>Are we supposed to make some sense of labels and try to come up with some models based on our data understanding? e.g. in Titanic use case we can build a model based  on theory that women will survive <br>\nOR we should ignore labels and loop though multiple models until we will find one with the best score? </p>",
  "messages": [
    {
      "id": "1327160",
      "postDate": "05/29/2021 02:24:48",
      "content": "<p>if it is 'anonymized, binary classification dataset ' what is the point of column labeling in the data set? </p>\n<p>Are we supposed to make some sense of labels and try to come up with some models based on our data understanding? e.g. in Titanic use case we can build a model based  on theory that women will survive <br>\nOR we should ignore labels and loop though multiple models until we will find one with the best score? </p>",
      "rawMarkdown": "if it is 'anonymized, binary classification dataset ' what is the point of column labeling in the data set? \n\nAre we supposed to make some sense of labels and try to come up with some models based on our data understanding? e.g. in Titanic use case we can build a model based  on theory that women will survive \nOR we should ignore labels and loop though multiple models until we will find one with the best score?",
      "votes": null
    },
    {
      "id": "1327163",
      "postDate": "05/29/2021 02:32:21",
      "content": "<p>Yes you can ignore the column names</p>",
      "rawMarkdown": "Yes you can ignore the column names",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1327163,
      "author_name": "nichaoku",
      "author_url": "",
      "post_date": "05/29/2021 02:32:21",
      "content": "<p>Yes you can ignore the column names</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1327160": "if it is 'anonymized, binary classification dataset ' what is the point of column labeling in the data set? \n\nAre we supposed to make some sense of labels and try to come up with some models based on our data understanding? e.g. in Titanic use case we can build a model based  on theory that women will survive \nOR we should ignore labels and loop though multiple models until we will find one with the best score?",
    "1327163": "Yes you can ignore the column names"
  },
  "source": "meta"
}