Online Machine Learning for Massive and Streaming Data

A Online machine learning framework which runs on top of the Apache Hama.

• Online SVM learning
• Online K-means clustering
• Online Bayesian learning
• Online Recommendation
• ..., etc.
  

Rationale and Algorithms

Work In Progress. I'm currently studying, and testing BSP applications that aim to speed up the recommendation engine for social commerce service.

Relationships with Other Apache Projects?

Comparing to Mahout, this is different in the sense that it runs based on Message-Passing paradigm and will have On-line processing features.