[Chicago] netflix prize

[gershon bialer]

If we were to go down that route, we could presumably get a copy of the
movie scripts, and do some natural language processing on the script. This
would probably only be useful if we didn't have that much rating data for
the script, but I suppose there could be some uses. Also, if the user really
likes certain actors, their presence in the movie could increase the rating.
However, I suppose should all be detectable straight from the data, assuming
there is sufficient data.

Gershon Bialer

On Thu, Mar 27, 2008 at 9:50 PM, Jon Sudlow <jsudlow at gmail.com> wrote:

> Your exactly right. I dont think netflix is looking for someone to crawl a
> third party data source to get a fuzzy match and use those results to help
> predict the ratings.
>
> By the way, when you do a group like this, one person is responsible to
> register the crew and communicate with Netflix. Who is the 'team leader',
> because we have to register to get our hands on the ratings data.
> -jon
>
>
> On Thu, Mar 27, 2008 at 5:53 PM, Kumar McMillan <kumar.mcmillan at gmail.com>
> wrote:
>
> > Massimo, I've attached the README from the download.  It explains
> > what's contained in the data and yes there is movie title and release
> > year.  I've also heard of how people have been getting sneaky and
> > matching the rating data (which just has CustomerID) to the movie
> > data, crawling IMDB, and "matching" netflix ratings to that of IMDB to
> > expose the identity behind CustomerIDs.  I don't think netflix was too
> > happy about this ;)  I just read it somewhere a while back, haven't
> > looked at the details.  Google gave me:
> > http://www.netflixprize.com/community/viewtopic.php?pid=5864
> >
> > On Thu, Mar 27, 2008 at 5:12 PM, Massimo Di Pierro
> > <mdipierro at cs.depaul.edu> wrote:
> > > Let us know what you find.
> > >
> > >  Massimo
> > >
> > >
> > >
> > >  On Mar 27, 2008, at 4:50 PM, Tom Printy wrote:
> > >
> > >  > On Thu, 2008-03-27 at 12:20 -0500, Massimo Di Pierro wrote:
> > >  >> Does anybody know what information is in the data they provide.
> > Other
> > >  >> than user's rating, what's in the movies database that one can use
> > >  >> for correlation (like actors in the movies, producers, genre,
> > etc.)?
> > >  > Well we could crawl IMDB for this info. Or I may be able to get my
> > >  > hands
> > >  > on the data ;)
> > >  >
> > >  >
> > >  > -Tom
> > >  >
> > >  > _______________________________________________
> > >  > Chicago mailing list
> > >  > Chicago at python.org
> > >  > http://mail.python.org/mailman/listinfo/chicago
> > >
> > >  _______________________________________________
> > >  Chicago mailing list
> > >  Chicago at python.org
> > >  http://mail.python.org/mailman/listinfo/chicago
> > >
> >
> > _______________________________________________
> > Chicago mailing list
> > Chicago at python.org
> > http://mail.python.org/mailman/listinfo/chicago
> >
> >
>
>
> --
> Jon Sudlow
> 3225 Foster Avenue
> 221 Sohlberg Hall
> C.P.O 2224
> Chicago, Il 60625
>
> _______________________________________________
> Chicago mailing list
> Chicago at python.org
> http://mail.python.org/mailman/listinfo/chicago
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/chicago/attachments/20080327/9a7f70c0/attachment.htm