Sunday 16 August 2015

mahout tutorial

      Introducing Mahout
      Recommendations
      First Recommender Engine
      Preference data
      PreferenceArray
      FastMap: Implementation of Map
      FastByIDMap
      FastIDSet
      DataModel
      GenericDataModel
      FileDataModel
      MySQLJDBCDataModel
      Generating Recommendations for Boolean data sets
      Item Similarity Vs User similarity
      User similarity
      CityBlockSimilarity : Compute User similarity
      EuclideanDistanceSimilarity
      PearsonCorrelationSimilarity : Compute User similarity
      SpearmanCorrelationSimilarity : Compute User similarity
      LogLikelihoodSimilarity : Compute User similarity
      CachingUserSimilarity: Compute User similarity
      UncenteredCosineSimilarity: Compute User similarity
      TanimotoCoefficientSimilarity: Compute User similarity
      Item similarity
      CityBlockSimilarity: Compute item similarity
      EuclideanDistanceSimilarity : Compute item similarity
      PearsonCorrelationSimilarity : Compute item similarity
      LogLikelihoodSimilarity : Compute item similarity
      UncenteredCosineSimilarity : Compute item similarity
      TanimotoCoefficientSimilarity : Compute item similarity
      CachingItemSimilarity : Compute item similarity
      UserNeighborhood
      NearestNUserNeighborhood
      ThresholdUserNeighborhood
      Recommender interface
      GenericUserBasedRecommender
      GenericBooleanPrefUserBasedRecommender
      SlopeOneRecommender
      CachingRecommender
      GenericItemBasedRecommender
      GenericBooleanPrefItemBasedRecommender
      ItemAverageRecommender
      ItemUserAverageRecommender
      SVDRecommender
      KnnItemBasedRecommender
      Cluster based recommendations

No comments:

Post a Comment