Mahout 算法列表

Collaborative Filtering

• User-Based Collaborative Filtering - single machine
• Item-Based Collaborative Filtering - single machine / MapReduce  【可以用多种相似度】
  
• Matrix Factorization with Alternating Least Squares - single machine /MapReduce
• Matrix Factorization with Alternating Least Squares on Implicit Feedback- single machine / MapReduce
• Weighted Matrix Factorization, SVD++, Parallel SGD - single machine

Classification

• Logistic Regression - trained via SGD - single machine   【SGD: Stochastic gradient descent 随机梯度下降】
  
• Naive Bayes/ Complementary Naive Bayes - MapReduce   【】
  
• Random Forest - MapReduce
• Hidden Markov Models - single machine
• Multilayer Perceptron - single machine

Clustering

• Canopy Clustering - single machine / MapReduce (deprecated, will be removed once Streaming k-Means is stable enough)
• k-Means Clustering - single machine / MapReduce
• Fuzzy k-Means - single machine / MapReduce
• Streaming k-Means - single machine / MapReduce
• Spectral Clustering - MapReduce

Dimensionality Reduction

• Singular Value Decomposition - single machine
• Lanczos Algorithm - single machine / MapReduce
• Stochastic SVD - single machine /MapReduce / Spark
• Principal Component Analysis (via Stochastic SVD)- single machine / MapReduce

Topic Models

• Latent Dirichlet Allocation - single machine / MapReduce

Miscellaneous

• Frequent Pattern Mining - MapReduce
• RowSimilarityJob - compute pairwise similarities between the rows of a matrix -MapReduce
• ConcatMatrices - combine 2 matrices or vectors into a single matrix - MapReduce
• Collocations - find co-locations of tokens in text - MapReduce