JAX London 2014: A retrospective
Scalable Library

Version 0.3 of Apache Mahout

Jessica Thornsby

Version 0.3 of the Apache Mahout library is now available.

Apache Mahout is a subproject of Apache Lucene and is built on Apache Hadoop. It aims to deliver a scalable, machine-learning algorithm implementation for settings like clustering, collaborative filtering and classification.

New for 0.3, are math and collections modules based on the Colt library, parallel Dirichlet process clustering, and a parallel co-occurrence based recommender. The command-line shell script has been updated to allow default Properties files. There is also a list of bug fixes. Please see the Release Notes for more information.

Author
Comments
comments powered by Disqus