{
  "id": 31548,
  "title": "9th place solution - Deep Breath",
  "content": "<p>You can read a post about our approach here:\n<a href=\"https://eliasvansteenkiste.github.io/machine%20learning/lung-cancer-pred/\">https://eliasvansteenkiste.github.io/machine%20learning/lung-cancer-pred/</a></p>\n\n<p>It is an initial version, we will update it with more details later.</p>",
  "comments": [
    {
      "id": 174901,
      "author_name": "ZheWang",
      "author_url": "",
      "post_date": "2017-04-13T06:20:34.217000",
      "content": "<p>Congrats! Looks like the Malignancy Prediction is the missing piece of our pipeline..We did not notice there is malignancy label in the LUNA dataset</p>",
      "votes": 3,
      "replies": [
        {
          "id": 174933,
          "author_name": "Fréderic Godin",
          "author_url": "",
          "post_date": "2017-04-13T08:29:35.780000",
          "content": "<p>Thanks and indeed.  Malignancy labels helped us a lot!</p>",
          "votes": 2,
          "replies": []
        }
      ]
    },
    {
      "id": 174894,
      "author_name": "zuchunshan",
      "author_url": "",
      "post_date": "2017-04-13T05:32:03.967000",
      "content": "<p>Congrats！Thanks very much for great sharing！I'm studying it.</p>",
      "votes": 2,
      "replies": []
    },
    {
      "id": 174887,
      "author_name": "Wiktor Macura",
      "author_url": "",
      "post_date": "2017-04-13T04:59:39.923000",
      "content": "<p>Fascinating! Thanks for the thorough writeup!</p>",
      "votes": 2,
      "replies": []
    },
    {
      "id": 178188,
      "author_name": "Oleg Trott",
      "author_url": "",
      "post_date": "2017-04-26T23:29:39.303000",
      "content": "<blockquote>\n  <p>Kaggle could easily prevent this in the future by truncating the\n  scores returned when submitting a set of predictions.</p>\n</blockquote>\n\n<p>I don't think that's accurate: If the leaderboard showed only 2 digits after the decimal point, that would make the top 5 results indistinguishable, but the \"perfect score script\" would still work -- it would just need about 10 days instead of 5.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 174926,
      "author_name": "Keyur K. Mithawala",
      "author_url": "",
      "post_date": "2017-04-13T07:52:57.073000",
      "content": "<p>Thank you Elias!</p>",
      "votes": 1,
      "replies": []
    }
  ],
  "comment_count": 6,
  "messages": [
    {
      "id": 174886,
      "postDate": "2017-04-13T04:52:57.737Z",
      "content": "<p>You can read a post about our approach here:\n<a href=\"https://eliasvansteenkiste.github.io/machine%20learning/lung-cancer-pred/\">https://eliasvansteenkiste.github.io/machine%20learning/lung-cancer-pred/</a></p>\n\n<p>It is an initial version, we will update it with more details later.</p>",
      "rawMarkdown": "You can read a post about our approach here:\nhttps://eliasvansteenkiste.github.io/machine%20learning/lung-cancer-pred/\n\nIt is an initial version, we will update it with more details later.",
      "votes": 33
    },
    {
      "id": 174901,
      "postDate": "2017-04-13T06:20:34.217Z",
      "content": "<p>Congrats! Looks like the Malignancy Prediction is the missing piece of our pipeline..We did not notice there is malignancy label in the LUNA dataset</p>",
      "rawMarkdown": "Congrats! Looks like the Malignancy Prediction is the missing piece of our pipeline..We did not notice there is malignancy label in the LUNA dataset",
      "votes": 3,
      "replies": [
        {
          "id": 174933,
          "postDate": "2017-04-13T08:29:35.780Z",
          "content": "<p>Thanks and indeed.  Malignancy labels helped us a lot!</p>",
          "rawMarkdown": "Thanks and indeed.  Malignancy labels helped us a lot!",
          "votes": 2
        }
      ]
    },
    {
      "id": 174894,
      "postDate": "2017-04-13T05:32:03.967Z",
      "content": "<p>Congrats！Thanks very much for great sharing！I'm studying it.</p>",
      "rawMarkdown": "Congrats！Thanks very much for great sharing！I'm studying it.",
      "votes": 2
    },
    {
      "id": 174887,
      "postDate": "2017-04-13T04:59:39.923Z",
      "content": "<p>Fascinating! Thanks for the thorough writeup!</p>",
      "rawMarkdown": "Fascinating! Thanks for the thorough writeup!",
      "votes": 2
    },
    {
      "id": 178188,
      "postDate": "2017-04-26T23:29:39.303Z",
      "content": "<blockquote>\n  <p>Kaggle could easily prevent this in the future by truncating the\n  scores returned when submitting a set of predictions.</p>\n</blockquote>\n\n<p>I don't think that's accurate: If the leaderboard showed only 2 digits after the decimal point, that would make the top 5 results indistinguishable, but the \"perfect score script\" would still work -- it would just need about 10 days instead of 5.</p>",
      "rawMarkdown": "> Kaggle could easily prevent this in the future by truncating the\n> scores returned when submitting a set of predictions.\n\nI don't think that's accurate: If the leaderboard showed only 2 digits after the decimal point, that would make the top 5 results indistinguishable, but the \"perfect score script\" would still work -- it would just need about 10 days instead of 5."
    },
    {
      "id": 174926,
      "postDate": "2017-04-13T07:52:57.073Z",
      "content": "<p>Thank you Elias!</p>",
      "rawMarkdown": "Thank you Elias!",
      "votes": 1
    }
  ],
  "complete": true,
  "source": "live",
  "fetched_at": "2026-10-24T13:31:54Z",
  "competition_id": 6004,
  "competition_slug": "data-science-bowl-2017",
  "team_id": 451175,
  "solution_url": "/competitions/data-science-bowl-2017/writeups/deep-breath-9th-place-solution-deep-breath"
}