You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@mahout.apache.org by ziad kamel <zi...@gmail.com> on 2012/03/28 23:45:23 UTC

How AveragingPreferenceInferrer works?

Hi  ,
It says "compute an inferred preference for a user and an item that
the user has not expressed any preference for".

Does it do it this way
1- infer
2- use it temporarily when checking  userSimilarity and choosing neighborhood
3-  then give a new preference ?

and in the wiki , I think "model" is missing
https://cwiki.apache.org/confluence/display/MAHOUT/Recommender+Documentation

check
userSimilarity.setPreferenceInferrer(new AveragingPreferenceInferrer());
to
userSimilarity.setPreferenceInferrer(new AveragingPreferenceInferrer(model));

Re: How AveragingPreferenceInferrer works?

Posted by Sean Owen <sr...@gmail.com>.
It pretends that any non-existent preference actually exists and is equal
to the user's average preference. It is only done for purposes of computing
similarity. It does not actually set a value in the model.

On Wed, Mar 28, 2012 at 10:45 PM, ziad kamel <zi...@gmail.com> wrote:

> Hi  ,
> It says "compute an inferred preference for a user and an item that
> the user has not expressed any preference for".
>
> Does it do it this way
> 1- infer
> 2- use it temporarily when checking  userSimilarity and choosing
> neighborhood
> 3-  then give a new preference ?
>
> and in the wiki , I think "model" is missing
>
> https://cwiki.apache.org/confluence/display/MAHOUT/Recommender+Documentation
>
> check
> userSimilarity.setPreferenceInferrer(new AveragingPreferenceInferrer());
> to
> userSimilarity.setPreferenceInferrer(new
> AveragingPreferenceInferrer(model));
>