{ "id": 6007, "title": "test data looks off", "url": "/competitions/yandex-personalized-web-search-challenge/discussion/6007", "author_name": "", "post_date": "2013-10-11T17:28:02.940Z", "votes": 1, "comment_count": 5, "views": 2365, "content": "

I haven't looked at the training data yet, but the test data seems formatted wrong for the "M" records.

\n

in the log file tab they are defined as

\n

SessionID Day TypeOfRecord USERID

\n

and an example is given like this

\n

744899 23 M 123123123

\n

but in the test data file I'm seeing

\n

34573630    M    28    15

\n

34573633    M    29    22

\n

34573634    M    28    27

\n

etc...

\n

 

\n

I haven't checked the training file yet but I wager there is a similar problem there.. we could compensate for the mistake in the data load but it sure would be easier if you guys changed your data :) (hopeful look)

", "messages": [ { "id": "32260", "postDate": "10/11/2013 17:28:02", "content": "

I haven't looked at the training data yet, but the test data seems formatted wrong for the "M" records.

\n

in the log file tab they are defined as

\n

SessionID Day TypeOfRecord USERID

\n

and an example is given like this

\n

744899 23 M 123123123

\n

but in the test data file I'm seeing

\n

34573630    M    28    15

\n

34573633    M    29    22

\n

34573634    M    28    27

\n

etc...

\n

 

\n

I haven't checked the training file yet but I wager there is a similar problem there.. we could compensate for the mistake in the data load but it sure would be easier if you guys changed your data :) (hopeful look)

", "rawMarkdown": "", "votes": null }, { "id": "32263", "postDate": "10/11/2013 18:05:34", "content": "

Hello,

\n

Thank you for the report! Indeed, there was a misalignment between the description and the dataset. We believe that at this point it is easier to fix the description.

", "rawMarkdown": "", "votes": null }, { "id": "32267", "postDate": "10/11/2013 19:44:32", "content": "

Either this is a mistake or I don't understand:

\n

[quote]These records describe the session (SessionID = 744899) of the user with USERID 123123123, performed on the 23rd day of the dataset. The user submitted the query with QUERYID 192902, which contains terms with TermIDs 4857,3847,2939. The URL with URLID 632428 placed on the domain DomainID 2384 is the top result on the corresponding SERP. 1403 units of time after beginning of the session the user clicked on the result with URLID 632428 (ranked second in the list).[/quote]

\n

On http://www.kaggle.com/c/yandex-personalized-web-search-challenge/details/logs-format

\n

Shouldn't that be: (ranked first in the list).

", "rawMarkdown": "", "votes": null }, { "id": "32273", "postDate": "10/11/2013 20:29:42", "content": "

Triskelion, yes, 632428 is the top ranked URLID. Thanks for noticing, we'll fix the example.

", "rawMarkdown": "", "votes": null }, { "id": "32828", "postDate": "10/29/2013 15:11:09", "content": "

In the test data set, why do some of the QueryAction records (i.e., lines with more than 5 fields) have TypeOfRecord = Q? There are 745,249 such lines.  797,867 of the QueryAction lines have TypeOfRecord = T which is the same as the number of Session metadata lines in that file.  According to the documentation, "T letter is used only for test queries."  I don't understand why the "Q" lines occur in the test file. 

", "rawMarkdown": "", "votes": null }, { "id": "32849", "postDate": "10/29/2013 18:23:45", "content": "

Test data contains not only test queries. There might be queries preceding the test one in a test session. They are not test queries and there is not need to re-rank documents for them. They are the context that you can use to re-rank documents for the test query.

", "rawMarkdown": "", "votes": null } ], "comments": [ { "id": 32263, "author_name": "eugene1751", "author_url": "", "post_date": "10/11/2013 18:05:34", "content": "

Hello,

\n

Thank you for the report! Indeed, there was a misalignment between the description and the dataset. We believe that at this point it is easier to fix the description.

", "votes": null, "replies": [] }, { "id": 32267, "author_name": "triskelion", "author_url": "", "post_date": "10/11/2013 19:44:32", "content": "

Either this is a mistake or I don't understand:

\n

[quote]These records describe the session (SessionID = 744899) of the user with USERID 123123123, performed on the 23rd day of the dataset. The user submitted the query with QUERYID 192902, which contains terms with TermIDs 4857,3847,2939. The URL with URLID 632428 placed on the domain DomainID 2384 is the top result on the corresponding SERP. 1403 units of time after beginning of the session the user clicked on the result with URLID 632428 (ranked second in the list).[/quote]

\n

On http://www.kaggle.com/c/yandex-personalized-web-search-challenge/details/logs-format

\n

Shouldn't that be: (ranked first in the list).

", "votes": null, "replies": [] }, { "id": 32273, "author_name": "eugene1751", "author_url": "", "post_date": "10/11/2013 20:29:42", "content": "

Triskelion, yes, 632428 is the top ranked URLID. Thanks for noticing, we'll fix the example.

", "votes": null, "replies": [] }, { "id": 32828, "author_name": "ralphw", "author_url": "", "post_date": "10/29/2013 15:11:09", "content": "

In the test data set, why do some of the QueryAction records (i.e., lines with more than 5 fields) have TypeOfRecord = Q? There are 745,249 such lines.  797,867 of the QueryAction lines have TypeOfRecord = T which is the same as the number of Session metadata lines in that file.  According to the documentation, "T letter is used only for test queries."  I don't understand why the "Q" lines occur in the test file. 

", "votes": null, "replies": [] }, { "id": 32849, "author_name": "serdyukovpv", "author_url": "", "post_date": "10/29/2013 18:23:45", "content": "

Test data contains not only test queries. There might be queries preceding the test one in a test session. They are not test queries and there is not need to re-rank documents for them. They are the context that you can use to re-rank documents for the test query.

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "32260": "", "32263": "", "32267": "", "32273": "", "32828": "", "32849": "" }, "source": "meta" }