mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From William <icswilliam2...@gmail.com>
Subject Re: parallelALS and RMSE TEST
Date Mon, 06 May 2013 09:17:06 GMT
Ted Dunning <ted.dunning <at> gmail.com> writes:

> 
> WIthout more information it is impossible to comment.
> 
> What experiments?
> 
> On Fri, May 3, 2013 at 8:45 AM, William <icswilliam2010 <at> gmail.com> 
wrote:
> 
> > I'm trying to get some recommendations with three Algorithms:
> > 1.parallelALS
> > 2.evaluateFactorization
> > 3.recommendfactorized
> >
> > In my experiments, RMSE value monotonically increases with larger
> > numfeatures.
> >
> > But Base on Netflix Prize experiment, RMSE should decreases with larger
> > numfeatures.
> >
> > How to explain and figure out it?
> >
> >
> >
> >
> 

I have a dataset about user and movie(no rate).But I want to get some 
recommendations from this dataset.
I just know the users see or not see some movie.So I set the rating matrix 
like: seen movies are 1, not seen movies are missing.
I use parallelALS function to decompose this matrix with three 
parameters(numfeatures ,numIterations, lambda). And I would like to get the 
best combination to reduce the RMSE.
I my experiment, RMSE value decreases with larger numIterations. But it 
increases with larger numfeatures. I use the another rating-matrix(from 
mahout official website) to experiment, everything is fine. 
So How to explain it? Can't I assign all rates are 1?
U  M  R
1,101,1
1,102,1
2,101,1
2,103,1
2,104,1
3,101,1
3,104,1
3,105,1
4,101,1
4,103,1





Mime
View raw message