A simple {@link org.apache.mahout.cf.taste.recommender.Recommender} which uses a given{@link org.apache.mahout.cf.taste.model.DataModel} and{@link org.apache.mahout.cf.taste.similarity.ItemSimilarity} to produce recommendations. This classrepresents Taste's support for item-based recommenders.
The {@link org.apache.mahout.cf.taste.similarity.ItemSimilarity} is the most important point to discusshere. Item-based recommenders are useful because they can take advantage of something to be very fast: they base their computations on item similarity, not user similarity, and item similarity is relatively static. It can be precomputed, instead of re-computed in real time.
Thus it's strongly recommended that you use {@link org.apache.mahout.cf.taste.impl.similarity.GenericItemSimilarity} with pre-computed similarities ifyou're going to use this class. You can use {@link org.apache.mahout.cf.taste.impl.similarity.PearsonCorrelationSimilarity} too, which computessimilarities in real-time, but will probably find this painfully slow for large amounts of data.
| |