[Xapian-devel] Is the project learning to rank need to be improved?

[Parth Gupta]

Dear Jiang,

You have downloaded correct data. The document available at

https://github.com/rishabhmehrotra/xapian/blob/master/xapian-letor/docs/letor.rst#how-to-use

will help you start with the code. Basically you need to index the
INEX collection with Omega (xapian-applications/omega/omindex.cc).
Just edit your omindex.cc as mentioned at the above link.  (e.g.
change in omindex.cc:  indexer.index_text(title, 5, "S"); ->
indexer.index_text(title, 1, "S"); )

Once the database (index) is created you can use the
xapian-letor-update.cc from xapain-letor/bin/ to update some index
statistics required by xapain-letor.

Then, use questletor.cc mentioning query-file and qrel-file from INEX.

Regards,
Parth.

On Tue, Oct 22, 2013 at 1:14 PM, jiangwen jiang <jiangwen127 at gmail.com> wrote:
> Hi, Parth:
>
> For xapian-letor, the test dataset is download from here?
> http://www.mpi-inf.mpg.de/departments/d5/software/inex/
>
> I downlaod this dataset: Official INEX 2009 collection.
>
> Can you give me a instruction to this dataset, is this dataset need to
> preprocess?
>
> Thanks
> Regards
>
>
> 2013/9/28 jiangwen jiang <jiangwen127 at gmail.com>
>>
>> Ok, I will try to start playing with the branch code.
>>
>> Regards
>>
>>
>> 2013/9/26 Parth Gupta <pargup8 at gmail.com>
>>>
>>> As Olly has already pointed out the 2012 branch is not merged.
>>>
>>> I think there are some compilation errors in the branch.
>>>
>>> The code in branch is better refactored. The Ranker and FeatureManager
>>> classes are well defined and implemented.
>>>
>>> Parth.
>>>
>>>
>>> On Wed, Sep 25, 2013 at 9:02 AM, Olly Betts <olly at survex.com> wrote:
>>>>
>>>> On Tue, Sep 24, 2013 at 08:34:10PM +0800, jiangwen jiang wrote:
>>>> > There's xapian-letor folder in current master trunk, is this code
>>>> > merged in
>>>> > 2011?
>>>>
>>>> Yes.
>>>>
>>>> > the letor code of 2012 is not merged to the trunk?
>>>>
>>>> It hasn't.  Sadly the 2012 project just tailed off.  I'm not sure what
>>>> the current state is compared to the code on trunk - Parth probably has
>>>> more of an idea.
>>>>
>>>> Cheers,
>>>>     Olly
>>>
>>>
>>
>