{
  "id": 10971,
  "title": "Global AUC-ROC scoring again?",
  "url": "/competitions/inria-bci-challenge/discussion/10971",
  "author_name": "",
  "post_date": "2014-11-20T00:06:26.977Z",
  "votes": 5,
  "comment_count": 1,
  "views": 1246,
  "content": "<p>The EEG competition that just closed used a global ROC metric so that the normalization of each patients score relative to the others mattered. Is that again the case here?</p>\n\n<p>Thanks</p>",
  "messages": [
    {
      "id": "58460",
      "postDate": "11/20/2014 00:06:26",
      "content": "<p>The EEG competition that just closed used a global ROC metric so that the normalization of each patients score relative to the others mattered. Is that again the case here?</p>\n\n<p>Thanks</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "58621",
      "postDate": "11/21/2014 16:43:09",
      "content": "<p>The aim is indeed to give a global ROC metric.</p>\n<p>SampleSubmission file is composed as follows :</p>\n<p>Each line&nbsp; should contain an Id (ex : S01_Sess01_FB001) and the prediction probabilty to detect good feedback (see Evaluation page).&nbsp;</p>\n\n<p>Therefore the aim is to compute ROC AUC over all subjects.</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 58621,
      "author_name": "manu13",
      "author_url": "",
      "post_date": "11/21/2014 16:43:09",
      "content": "<p>The aim is indeed to give a global ROC metric.</p>\n<p>SampleSubmission file is composed as follows :</p>\n<p>Each line&nbsp; should contain an Id (ex : S01_Sess01_FB001) and the prediction probabilty to detect good feedback (see Evaluation page).&nbsp;</p>\n\n<p>Therefore the aim is to compute ROC AUC over all subjects.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "58460": "",
    "58621": ""
  },
  "source": "meta"
}