That's strange, since I've run the same data set and never seen behavior like this. Yes I run on my laptop too, which is fairly similar.
Yes of course the time is consumed somewhere from recommend(), but where? I think you'd want to get some clue about where within this processing the time is being consumed. 2010/7/14 Young <[email protected]>: > I tried the 10M dataset from grouplen. Is it the reason I do the project in > my own laptop? It is Intel 2 core 2.4 GB, and RAM 3GB, and Win7 OS. > And blow is profiled code. > ----------------------------------------------------- > //Precompute the model, itemSimilarity. > DataModel model = new GroupLensDataModel(new File("ratings.dat")); > ItemSimilarity itemSimilarity = null; > try { > itemSimilarity = new PearsonCorrelationSimilarity(model); > } catch (TasteException e) { > e.printStackTrace(); > } > Recommender recommender = new > GenericItemBasedRecommender(model,itemSimilarity); > ----------------------------------------- > > //Below method consume more than 1min to generate result. > itembased_items = recommender.recommend(user_id, 10); > -------------------------------------------------- > Should I try slope-one? > > > > >
