Date: Wednesday, 10 October 2007, 6:30 PM
Location: SAP LABS, Building D, 3410 Hillview Avenue,
Palo Alto, CA (Google
Maps | Yahoo!
Maps | Mapquest)
Cost: Free and open to all who wish to attend, but membership is
only $10/year.
Topic
Several approaches to collaborative filtering have been studied but seldom have the studies been reported for large (several millions of users and items) and dynamic (the underlying item set is continually changing) settings. This talk will focus on our approach to collaborative filtering for generating personalized recommendations for users of Google News. We generate recommendations using three approaches: collaborative filtering using MinHash clustering, Probabilistic Latent Semantic Indexing (PLSI), and covisitation counts. We combine recommendations from different algorithms using a linear model. Our approach is content agnostic and consequently domain independent, making it easily adaptible for other applications and languages with minimal effort. The talk will describe our algorithms and system setup in detail, and report results of running the recommendations engine on Google News.
About the Speaker
Mayur Datar works as a Research Scientist with Google Inc. His research interests are in datamining, algorithms, databases and computer science theory. Prior to joining Google, Mayur obtained his doctorate degree in computer science from Stanford university and a Bachelor of Technology degree from I.I.T. Bombay. He was awarded the President of India, Gold Medal for being the most outstanding student of his graduating batch from I.I.T. Bombay. He has published several papers in renowned conferences such as SIGMOD, VLDB, KDD, FOCS, SODA, WWW.