Mahout: How to user IDRescorer in Distributed mode.?

[#1]

I have worked with ID Rescorer and the Recommendation in standalone mode. But, is there a way that we can achieve the similar process in Distributed mode as well.?

The simiilarity classes in Distributed mode work in different manner, as every one of them will extend VectorSimilarityMeasure and there wont be any method as recommend as such.

[#2]

No, IDRescorer has always been a part only of the non-distributed implementation.

[#3]

Thanks Sean.

I want one more suggestion from you.

I want to provide recommendations based on user profile, and item data that too considering various features. for eg:, If a user purchases and rates a book which is of french language and of thriller genre. So, out of the recommendations i got, i need to boost french & thriller books first. 

i am thinking few options, one is clustering based recommendation which clusters data according to genre or language etc.

second one is to, plug the search engine after the recommendations. will be glad if you can suggest a way ahead.

Also, does the ALS Factorizer on Implicit data peforms recommendataion based on ratings and user features as well.?

[#4 · Accepted solution]

That's really what IDRescorer is for, yes. If you need it in distributed mode you can reimplement the same idea by changing the code. I don't think it's really a clustering problem; you're just filtering based on clear attributes. You could also think of it a search relevance problem, and combine the results of a recommender and search engine in your app. No, ALS has no concept of attributes. It's a different, longer story, but you can always use 'fake' users and items corresponding to topics or labels to inject this information in the ALS model.