Big Data Bellevue Meetup - Recommendations Secrets: How to build a multi-modal recommender
Bellevue, WA
Wednesday, May 20, 2015
The Big Data Bellevue Meetup is a community for everyone interested in big data around the Bellevue area. This is a place for anyone, technical or business, to share your success stories and best practices about Big Data.


Recommendations Secrets: How to build a multi-modal recommender

Joseph Blue View Bio

This talk will cover: 1. How to build a production quality recommendation engine using a search engine and Mahout 2. How to build a multi-modal recommendation from multiple behavioral inputs 3. How search engines can be used for more than just text To do this, we will do detailed tear-down and walk-through of a working soup-to-nuts recommendation engine that uses observations of multiple kinds of behavior to do combined recommendation and cross recommendation. The system is built using Mahout to do off-line analysis and Solr to provide real-time recommendations. The talk will also include enough theory to provide useful working intuitions for those desiring to adapt this design. Building recommendation engines by abusing a search engine has been well-known for some time to a small sub-culture in the recommendation community, but techniques for building multi-modal recommendation engines are not at all well known.


Joseph Blue

In his role as Data Scientist at MapR, Joe assists customers in solving their big data problems, making efficient use of the Hadoop ecosystem to generate tangible results. Recent projects include debit card fraud & breach detection, lead generation from social data, customer matching through record linkage, lookalike modeling using browser history and real-time product recommendations.

Prior to MapR, Joe was the Chief Scientist for Optum (a division of UnitedHealth) and the principal innovator in analytics for healthcare. As a Sr. Fellow with OptumLabs, he applied machine learning concepts to healthcare issues such as disease prediction from co-morbidities, estimation of PMPY (member cost), physician scoring and treatment pathways. As a leader in the Payment Integrity business, he built anomaly detection engines responsible for saving $100M annually in claim overpayments.