Package org.apache.mahout.cf.taste.hadoop.item

Class Summary
AggregateAndRecommendReducer computes prediction values for each user
ItemFilterAsVectorAndPrefsReducer we use a neat little trick to explicitly filter items for some users: we inject a NaN summand into the preference estimation for those items, which makes AggregateAndRecommendReducer automatically exclude them
ItemFilterMapper map out all user/item pairs to filter, keyed by the itemID
ItemIDIndexMapper  
ItemIDIndexReducer  
PartialMultiplyMapper maps similar items and their preference values per user
PrefAndSimilarityColumnWritable  
RecommenderJob Runs a completely distributed recommender job as a series of mapreduces.
SimilarityMatrixRowWrapperMapper maps a row of the similarity matrix to a VectorOrPrefWritable actually a column from that matrix has to be used but as the similarity matrix is symmetric, we can use a row instead of having to transpose it
ToUserVectorReducer Input
ToVectorAndPrefReducer  
UserVectorSplitterMapper  
VectorAndPrefsWritable  
VectorOrPrefWritable  
 



Copyright © 2008-2010 The Apache Software Foundation. All Rights Reserved.