{
  "id": 6302,
  "title": "Tips for beginners",
  "url": "/competitions/accelerometer-biometric-competition/discussion/6302",
  "author_name": "r0u1i",
  "post_date": "2013-11-13T07:05:42.847000",
  "votes": 3,
  "comment_count": 0,
  "views": 1733,
  "content": "<p>I was approached via email by one of the students participating in this competition, asking for tips. I don't think my answer was that helpful, but for fairness' sake, I'm publishing it here:</p>\n<p>&nbsp;</p>\n<p>&quot;</p>\n<p>I started this competition to learn how to approach time series problems like the one in the core of this challenge.<br>As my group name suggests, I found out that you can ignore the &quot;data&quot;, and find other tricks that can get you quite far on the leaderboard.</p>\n<p>Ask yourself what features can identify a user, other than averaging his/her values, and then test those features. Average is just a single &quot;feature&quot;. <br>Can the variance in values tell us anything?<br>How about the time of day (maybe users come from different time zones)?<br>Are there unique x,y,z sequences for each user?</p>\n<p>One more tip - Bayes theorem rocks.</p>\n<p>&quot;</p>\n<p>&nbsp;</p>",
  "messages": [
    {
      "id": 33663,
      "postDate": "2013-11-13T07:05:42.847Z",
      "content": "<p>I was approached via email by one of the students participating in this competition, asking for tips. I don't think my answer was that helpful, but for fairness' sake, I'm publishing it here:</p>\n<p>&nbsp;</p>\n<p>&quot;</p>\n<p>I started this competition to learn how to approach time series problems like the one in the core of this challenge.<br>As my group name suggests, I found out that you can ignore the &quot;data&quot;, and find other tricks that can get you quite far on the leaderboard.</p>\n<p>Ask yourself what features can identify a user, other than averaging his/her values, and then test those features. Average is just a single &quot;feature&quot;. <br>Can the variance in values tell us anything?<br>How about the time of day (maybe users come from different time zones)?<br>Are there unique x,y,z sequences for each user?</p>\n<p>One more tip - Bayes theorem rocks.</p>\n<p>&quot;</p>\n<p>&nbsp;</p>",
      "votes": 3
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "33663": ""
  }
}