Mahout: Item similarity

Item-item models use rating distributions per item, not per user. Suppose there are more users than items, each item tends to have more ratings than each user, so an item's average rating usually doesn't change quickly.

Mahout provides following implementations for item-to-item collaborative filtering. Following classes implements ItemSimilarity interface. ItemSimilarity finds similarity between two items.

a.      CachingItemSimilarity

b.      CityBlockSimilarity

c.       EuclideanDistanceSimilarity

d.      FileItemSimilarity

e.      GenericItemSimilarity

f.       LogLikelihoodSimilarity

g.      PearsonCorrelationSimilarity

h.      TanimotoCoefficientSimilarity
i.       UncenteredCosineSimilarity

Prevoius                                                 Next                                                 Home