{
  "id": 6625,
  "title": "Time complexity and exp cost",
  "url": "/competitions/facebook-recruiting-iii-keyword-extraction/discussion/6625",
  "author_name": "",
  "post_date": "2013-12-18T10:06:42.587Z",
  "votes": null,
  "comment_count": 2,
  "views": 1714,
  "content": "<p>hi there,</p>\n<p>I have 2 main questions and want to describe the scenario before asking</p>\n<p>I implemented a prototype solving this issue from an online methodology:<br>http://php.scripts.psu.edu/users/s/z/szt5115/publications/2013_fp050-tuarob.pdf<br>For short, this is a big O(n^3) complexity algorithm. I have to prune them such as sampling in order to make a submission before deadline or wait about several years :S<br>However, the F1 is totally worse, that is, I make a diseconomy spending about 700 USD for running mapreduce through AWS EMR for experiment and get the bad result.</p>\n<p>So, here is my question:</p>\n<p>1. How can I achieve better score using lower complexity algorithm toward this competition? Could someone give me some hint about your methodology?</p>\n<p>2. How much money and duration in average you cost in this competition per experiment to gain the score higher than 0.70?</p>\n<p>thanks in advance</p>",
  "messages": [
    {
      "id": "36294",
      "postDate": "12/18/2013 10:06:42",
      "content": "<p>hi there,</p>\n<p>I have 2 main questions and want to describe the scenario before asking</p>\n<p>I implemented a prototype solving this issue from an online methodology:<br>http://php.scripts.psu.edu/users/s/z/szt5115/publications/2013_fp050-tuarob.pdf<br>For short, this is a big O(n^3) complexity algorithm. I have to prune them such as sampling in order to make a submission before deadline or wait about several years :S<br>However, the F1 is totally worse, that is, I make a diseconomy spending about 700 USD for running mapreduce through AWS EMR for experiment and get the bad result.</p>\n<p>So, here is my question:</p>\n<p>1. How can I achieve better score using lower complexity algorithm toward this competition? Could someone give me some hint about your methodology?</p>\n<p>2. How much money and duration in average you cost in this competition per experiment to gain the score higher than 0.70?</p>\n<p>thanks in advance</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "36303",
      "postDate": "12/18/2013 11:45:00",
      "content": "<p>1. Read <a href=\"http://www.kaggle.com/c/facebook-recruiting-iii-keyword-extraction/forums/t/6416/new-baseline\">this</a>:&nbsp;http://www.kaggle.com/c/facebook-recruiting-iii-keyword-extraction/forums/t/6416/new-baseline . Based on the size of the dataset, nothing but an O(n) algorithm would be suitable. Take a look at how the first thousand post and see how you might guess the tags then implement it as your tag predictor.</p>\n<p>2. $0(not including cost of electricity and opportunity cost).</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "36367",
      "postDate": "12/19/2013 14:41:48",
      "content": "<p>hi Mr. Grant,</p>\n<p>&nbsp;</p>\n<p>your observation is truly work!! infect, i didn't even take a glance at the data such that didn't find any of the obviously issues (e.g., duplicate entry in this case)</p>\n<p>&nbsp;</p>\n<p>Thank you for help =)</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 36303,
      "author_name": "chestergrant",
      "author_url": "",
      "post_date": "12/18/2013 11:45:00",
      "content": "<p>1. Read <a href=\"http://www.kaggle.com/c/facebook-recruiting-iii-keyword-extraction/forums/t/6416/new-baseline\">this</a>:&nbsp;http://www.kaggle.com/c/facebook-recruiting-iii-keyword-extraction/forums/t/6416/new-baseline . Based on the size of the dataset, nothing but an O(n) algorithm would be suitable. Take a look at how the first thousand post and see how you might guess the tags then implement it as your tag predictor.</p>\n<p>2. $0(not including cost of electricity and opportunity cost).</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 36367,
      "author_name": "theringe",
      "author_url": "",
      "post_date": "12/19/2013 14:41:48",
      "content": "<p>hi Mr. Grant,</p>\n<p>&nbsp;</p>\n<p>your observation is truly work!! infect, i didn't even take a glance at the data such that didn't find any of the obviously issues (e.g., duplicate entry in this case)</p>\n<p>&nbsp;</p>\n<p>Thank you for help =)</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "36294": "",
    "36303": "",
    "36367": ""
  },
  "source": "meta"
}