mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Dinesh B Vadhia" <>
Subject Re: Two learning competitions that might be of interest for Mahout
Date Tue, 15 Feb 2011 15:54:13 GMT
Yep, weird that it is ratings only.  Better still to have both ratings and playcounts.  

From: Ted Dunning 
Sent: Tuesday, February 15, 2011 7:41 AM
Subject: Re: Two learning competitions that might be of interest for Mahout

For music, there is an even bigger difference between ratings and what
people want to listen to.  It is, indeed, a pity that the data is ratings
instead of listening histories.

On Tue, Feb 15, 2011 at 3:57 AM, Sean Owen <> wrote:

> If I may guess at the answer --
> Yes in theory it would be better to score output on the quality of its
> top recommendations, rather than on accuracy of predicted ratings,
> which are just one means to that goal. There are of course contexts
> where you have no ratings, so the winning technique here may not
> translate to those scenarios.
> Perhaps output would be scored on what proportion of the top k match
> the real top k preferred items. And so the test would actually
> withhold the top k rated items and ask recommenders to predict them.
> This has two problems I can see, however.
> The small problem is that chopping off the top ratings makes the test
> data systematically different than real data. There's a lot of
> "information" in those top ratings versus any arbitrary k.
> The bigger problem is that the user's top k ratings are not
> necessarily the same as the best k recommendations! Let's say I've
> never seen the movie Breathless, but, if I do, I'll find it's actually
> my favorite movie ever. A recommender would be right in making this a
> top recommendation. But a recommender evaluation framework such as
> this contest might use can't know that, so would count that "wrong".
> Evaluating rating accuracy is at least unambiguous in comparison and
> so can form the basis of a competition.
> And to be fair, most people making production recommender systems
> would expect it to be able to estimate a rating, in addition to making
> recommendations.
> On Tue, Feb 15, 2011 at 11:19 AM, Chen_1st <> wrote:
> > Hi, Markus,
> >
> > I am curious why the competition still tries to predict the rating
> > values, now that top k recommendation is more practical in real life
> > applications, and it's illustrated by many papers that rating value
> > prediction is not so useful for discovery of top k items.
> >
> > Best Regards.
> >
> > Chen

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message