{
  "id": 7908,
  "title": "0.59410",
  "url": "/competitions/acquire-valued-shoppers-challenge/discussion/7908",
  "author_name": "",
  "post_date": "2014-04-28T15:54:24.570Z",
  "votes": null,
  "comment_count": 6,
  "views": 2719,
  "content": "<p>There's a lot of people on the leaderboard with an AUC of&nbsp;0.59347, which makes sense, since&nbsp;<a href=\"https://www.kaggle.com/c/acquire-valued-shoppers-challenge/forums/t/7688/feature-engineering-and-beat-the-benchmark-0-59347\">Triskelion posted example code</a>&nbsp;that achieves that level of accuracy.</p>\n<p>However, even more people have an exact score of&nbsp;0.59410. What do those people know that the rest of us do not? &nbsp;Is there some easy improvement to Triskelion's model that a bunch of people arrived at independently?</p>",
  "messages": [
    {
      "id": "43222",
      "postDate": "04/28/2014 15:54:24",
      "content": "<p>There's a lot of people on the leaderboard with an AUC of&nbsp;0.59347, which makes sense, since&nbsp;<a href=\"https://www.kaggle.com/c/acquire-valued-shoppers-challenge/forums/t/7688/feature-engineering-and-beat-the-benchmark-0-59347\">Triskelion posted example code</a>&nbsp;that achieves that level of accuracy.</p>\n<p>However, even more people have an exact score of&nbsp;0.59410. What do those people know that the rest of us do not? &nbsp;Is there some easy improvement to Triskelion's model that a bunch of people arrived at independently?</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "43223",
      "postDate": "04/28/2014 16:11:08",
      "content": "<p>They used latest version of vowpal-wabbit</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "43224",
      "postDate": "04/28/2014 16:13:37",
      "content": "<p>Really? &nbsp;I used 7.6, and actually got a slightly worse score than&nbsp;0.59347</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "43310",
      "postDate": "04/29/2014 16:33:12",
      "content": "<p>Ok, I had a typo in my code. &nbsp;When I use Vowpal Wabbit 7.6 on the code&nbsp;Triskelion posted, I do indeed get a leaderboard AUC of exactly&nbsp;0.59410</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "43311",
      "postDate": "04/29/2014 16:39:50",
      "content": "<p>See. I told you :P</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "43312",
      "postDate": "04/29/2014 16:40:44",
      "content": "<p>Hah, you did indeed.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "43313",
      "postDate": "04/29/2014 16:41:46",
      "content": "<p>It's interesting-- running VW in logistic regression model, with the exact same dataset, gives an AUC of ~0.55. &nbsp;(You have to change the target from 0/1 to -1/1).</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 43223,
      "author_name": "abhishek",
      "author_url": "",
      "post_date": "04/28/2014 16:11:08",
      "content": "<p>They used latest version of vowpal-wabbit</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 43224,
      "author_name": "zachmayer",
      "author_url": "",
      "post_date": "04/28/2014 16:13:37",
      "content": "<p>Really? &nbsp;I used 7.6, and actually got a slightly worse score than&nbsp;0.59347</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 43310,
      "author_name": "zachmayer",
      "author_url": "",
      "post_date": "04/29/2014 16:33:12",
      "content": "<p>Ok, I had a typo in my code. &nbsp;When I use Vowpal Wabbit 7.6 on the code&nbsp;Triskelion posted, I do indeed get a leaderboard AUC of exactly&nbsp;0.59410</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 43311,
      "author_name": "abhishek",
      "author_url": "",
      "post_date": "04/29/2014 16:39:50",
      "content": "<p>See. I told you :P</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 43312,
      "author_name": "zachmayer",
      "author_url": "",
      "post_date": "04/29/2014 16:40:44",
      "content": "<p>Hah, you did indeed.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 43313,
      "author_name": "zachmayer",
      "author_url": "",
      "post_date": "04/29/2014 16:41:46",
      "content": "<p>It's interesting-- running VW in logistic regression model, with the exact same dataset, gives an AUC of ~0.55. &nbsp;(You have to change the target from 0/1 to -1/1).</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "43222": "",
    "43223": "",
    "43224": "",
    "43310": "",
    "43311": "",
    "43312": "",
    "43313": ""
  },
  "source": "meta"
}