[Corpora-List] Recommending systems training data

Max Chevalier Max.Chevalier at irit.fr
Tue Mar 18 14:12:56 CET 2014


Le 18/03/2014 13:18, Ivelina Nikolova a écrit :
> Dear Corpora-members,
>
> I'm looking for freely available data for training recommending
> systems - some resource where multiple users and their choices of
> articles over time are available. My task is to model the user
> behavior and offer further reads based on the history of his reads.
>
> I asked few weeks ago about corpora on document similarity and
> received valuable feedback, but this task includes also some user
> behavior modeling and I wonder if there is some golden data especially
> for recommending systems. I would appreciate any pointers.
>
> Thanks in advance!
> Ivelina
>

Dear Ivelina,

What about the famous MovieLens collection? I concerns films but you have a lot of data. Do you need the content of recommended objects?

Best regards,

Max Chevalier -------------- next part -------------- A non-text attachment was scrubbed... Name: Max_Chevalier.vcf Type: text/x-vcard Size: 366 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20140318/2fbb2530/attachment.vcf>



More information about the Corpora mailing list