A Mahout is a person who tends to elephants. Thus, the Apache Hadoop project—which has an elephant for a mascot—is now being tended by version 0.8 of the Mahout machine-learning libraries. Mahout 0.8 is primarily a bug fix and performance-improvement release, focusing on faster vector and matrix handling, and improved recommendations.
Those recommendations are of the “if user likes A, they might also like B, C and D,” variety. In previous releases, Mahout’s recommendation implementations were able to create such recommendation systems based on Hadoop-stored user data. Mahout 0.8 includes refinements to this system to allow such recommendations to be generated faster and more reliably.
According to the Apache Foundation, version 0.9 of Mahout will continue to address bug fixes and performance improvements. From the Apache Foundation: “As the project moves toward a 1.0 release, the community is working to clean up and/or remove parts of the codebase that are under-supported or that underperform as well as to better focus the energy and contributions on key algorithms that are proven to scale in production and have seen widespread adoption.”
