Featured Author

Pat Ferrel
Principal Consultant, FinderBots

Pat is principal consultant at FinderBots, a consultancy specializing in the delivery of applications of machine learning including recommenders and big-data predictive analytics.  He is also a committer to Apache Mahout where he works on the next generation of recommenders.  Visit his consulting service at Finderbots here.  


Author's Posts

Posted on September 5, 2014 by Pat Farrel
Combining a search engine with Mahout has created a recommender that is extremely fast and scalable and seamlessly blends results using collaborative filtering data and metadata. In the first post we described creating a co-occurrence indicator matrix for a recommender. In this follow up post, we dive in deeper to the performance and quality of the recommendations.
Posted on August 12, 2014 by Pat Farrel
There are big changes happening in Apache Mahout. For years it’s been the go to machine learning library for Hadoop. It contained most of the best-in-class algorithms for scalable machine learning, which means clustering, classification, and recommendation. But it was written for Hadoop and mapreduce. Today a number of new parallel execution engines show great promise in speeding calculations by as much as 10-100x (Spark, H2O, Flink). That means instead of buying 10 computers for a cluster, a single one may do. That should get you manager’s attention.

Blog Sign Up

Sign up and get the top posts from each week delivered to your inbox every Friday!