Yes, you want the sampling one so you can reduce the number of
neighbors you consider.
On Fri, May 11, 2012 at 6:47 PM, Emilio Suarez <Emilio.Suarez@intela.com> wrote:
> Thanks Sean,
>
> So, do you suggest something like this?
>
> LogLikelihoodSimilarity similarity = new LogLikelihoodSimilarity(fileDataModel);
> PreferredItemsNeighborhoodCandidateItemsStrategy candidateStrategy = new PreferredItemsNeighborhoodCandidateItemsStrategy();
> recommender = new GenericItemBasedRecommender(fileDataModel, similarity, candidateStrategy,
candidateStrategy);
>
> or this?
>
> LogLikelihoodSimilarity similarity = new LogLikelihoodSimilarity(fileDataModel);
> SamplingCandidateItemsStrategy candidateStrategy = new SamplingCandidateItemsStrategy();
> recommender = new GenericItemBasedRecommender(fileDataModel, similarity, candidateStrategy,
candidateStrategy);
>
>
> -emilio
>
> You need to apply a CandidateItemStrategy to reduce the number of
> elements you consider, or else it will take a very long time because
> almost the entire model is a candidate for recommendation.
>
|