Mahout 0.12.0 Features by Engine
—

  Single Machine MapReduce Spark H2O Flink
Mahout Math-Scala Core Library and Scala DSL          
Mahout Distributed BLAS. Distributed Row Matrix API with R and Matlab like operators. Distributed ALS, SPCA, SSVD, thin-QR. Similarity Analysis.     x x x
           
Mahout Interactive Shell          
Interactive REPL shell for Spark optimized Mahout DSL     x    
           
Collaborative Filtering with CLI drivers          
User-Based Collaborative Filtering deprecated deprecated x    
Item-Based Collaborative Filtering x x x    
Matrix Factorization with ALS x x      
Matrix Factorization with ALS on Implicit Feedback x x      
Weighted Matrix Factorization, SVD++ x        
           
Classification with CLI drivers          
Logistic Regression - trained via SGD deprecated        
Naive Bayes / Complementary Naive Bayes   deprecated x    
Hidden Markov Models deprecated        
           
Clustering with CLI drivers          
Canopy Clustering deprecated deprecated      
k-Means Clustering deprecated deprecated      
Fuzzy k-Means deprecated deprecated      
Streaming k-Means deprecated deprecated      
Spectral Clustering   deprecated      
           
Dimensionality Reduction note: most scala-based dimensionality reduction algorithms are available through the Mahout Math-Scala Core Library for all engines          
Singular Value Decomposition deprecated deprecated x x x
Lanczos Algorithm deprecated deprecated      
Stochastic SVD deprecated deprecated x x x
PCA (via Stochastic SVD) deprecated deprecated x x x
QR Decomposition deprecated deprecated x x x
           
Topic Models          
Latent Dirichlet Allocation deprecated deprecated      
           
Miscellaneous          
RowSimilarityJob   deprecated x    
Collocations   deprecated      
Sparse TF-IDF Vectors from Text   deprecated      
XML Parsing   deprecated      
Email Archive Parsing   deprecated      
Evolutionary Processes x