{
  "id": 73540,
  "title": "Understanding train.csv",
  "url": "/competitions/elo-merchant-category-recommendation/discussion/73540",
  "author_name": "",
  "post_date": "2018-12-03T23:43:50.887739700Z",
  "votes": null,
  "comment_count": 4,
  "views": 0,
  "content": "<p>Hi,</p>\n\n<p>I wanted to understand what do these variables mean in the training.csv:\n1. feature_1\n2. feature_2\n3. feature_3</p>\n\n<p>Thanks,\nAkshay Kher</p>",
  "messages": [
    {
      "id": "432480",
      "postDate": "12/03/2018 23:43:50",
      "content": "<p>Hi,</p>\n\n<p>I wanted to understand what do these variables mean in the training.csv:\n1. feature_1\n2. feature_2\n3. feature_3</p>\n\n<p>Thanks,\nAkshay Kher</p>",
      "rawMarkdown": "Hi,\n\nI wanted to understand what do these variables mean in the training.csv:\n1. feature_1\n2. feature_2\n3. feature_3\n\nThanks,\nAkshay Kher",
      "votes": null
    },
    {
      "id": "432565",
      "postDate": "12/04/2018 03:41:19",
      "content": "<p>They are Anonymized, so we have no way of knowing what do they mean.</p>",
      "rawMarkdown": "They are Anonymized, so we have no way of knowing what do they mean.",
      "votes": null
    },
    {
      "id": "432683",
      "postDate": "12/04/2018 07:20:46",
      "content": "<p>Hi Akshay,</p>\n\n<p>The ELO dataset comes with anonymized data and they have performed some operations on the data to keep up with the Privacy Compliance and Assurance. But it is pretty easy to understand the data by performing some data preprocessing and EDA to find the correlation among the data points. Generally we are bound to find some correlation or data leaks performing EDA. </p>\n\n<p>You can checkout <a href=\"https://www.youtube.com/watch?v=W5WE9Db2RLU\">this PyCon video</a> that shows some live coding for EDA. Happy coding :)</p>",
      "rawMarkdown": "Hi Akshay,\n\nThe ELO dataset comes with anonymized data and they have performed some operations on the data to keep up with the Privacy Compliance and Assurance. But it is pretty easy to understand the data by performing some data preprocessing and EDA to find the correlation among the data points. Generally we are bound to find some correlation or data leaks performing EDA. \n\nYou can checkout [this PyCon video](https://www.youtube.com/watch?v=W5WE9Db2RLU) that shows some live coding for EDA. Happy coding :)",
      "votes": null
    },
    {
      "id": "434003",
      "postDate": "12/05/2018 18:51:39",
      "content": "<p>Moreover, the whole data set is simulated.</p>",
      "rawMarkdown": "Moreover, the whole data set is simulated.",
      "votes": null
    },
    {
      "id": "434921",
      "postDate": "12/07/2018 06:53:12",
      "content": "<p>Some simulated features, maybe produced by their machine learning algorithms. You can find explanations for each feature in the Data_Dictionary.xlsx, provided in \"Data\".</p>",
      "rawMarkdown": "Some simulated features, maybe produced by their machine learning algorithms. You can find explanations for each feature in the Data_Dictionary.xlsx, provided in \"Data\".",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 432565,
      "author_name": "artgor",
      "author_url": "",
      "post_date": "12/04/2018 03:41:19",
      "content": "<p>They are Anonymized, so we have no way of knowing what do they mean.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 432683,
      "author_name": "thanatoz",
      "author_url": "",
      "post_date": "12/04/2018 07:20:46",
      "content": "<p>Hi Akshay,</p>\n\n<p>The ELO dataset comes with anonymized data and they have performed some operations on the data to keep up with the Privacy Compliance and Assurance. But it is pretty easy to understand the data by performing some data preprocessing and EDA to find the correlation among the data points. Generally we are bound to find some correlation or data leaks performing EDA. </p>\n\n<p>You can checkout <a href=\"https://www.youtube.com/watch?v=W5WE9Db2RLU\">this PyCon video</a> that shows some live coding for EDA. Happy coding :)</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 434003,
      "author_name": "kailex",
      "author_url": "",
      "post_date": "12/05/2018 18:51:39",
      "content": "<p>Moreover, the whole data set is simulated.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 434921,
      "author_name": "salamatburj",
      "author_url": "",
      "post_date": "12/07/2018 06:53:12",
      "content": "<p>Some simulated features, maybe produced by their machine learning algorithms. You can find explanations for each feature in the Data_Dictionary.xlsx, provided in \"Data\".</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "432480": "Hi,\n\nI wanted to understand what do these variables mean in the training.csv:\n1. feature_1\n2. feature_2\n3. feature_3\n\nThanks,\nAkshay Kher",
    "432565": "They are Anonymized, so we have no way of knowing what do they mean.",
    "432683": "Hi Akshay,\n\nThe ELO dataset comes with anonymized data and they have performed some operations on the data to keep up with the Privacy Compliance and Assurance. But it is pretty easy to understand the data by performing some data preprocessing and EDA to find the correlation among the data points. Generally we are bound to find some correlation or data leaks performing EDA. \n\nYou can checkout [this PyCon video](https://www.youtube.com/watch?v=W5WE9Db2RLU) that shows some live coding for EDA. Happy coding :)",
    "434003": "Moreover, the whole data set is simulated.",
    "434921": "Some simulated features, maybe produced by their machine learning algorithms. You can find explanations for each feature in the Data_Dictionary.xlsx, provided in \"Data\"."
  },
  "source": "meta"
}