AIC Seminar Series

I will describe the TiVo TV-show collaborative recommendation system which is fielded in over three million TiVos for seven years. Over this install base, TiVo has more than 100 million ratings of approximately 30,000 distinct TV shows and movies. TiVo uses an item-item form of collaborative filtering which preserves privacy so obviating the need to keep a memory on the server for each TiVo user’s viewing preferences. Despite the high profile nature of this collaborative filtering system, because of this strong privacy protection approach, TiVo has suffered no privacy backlash. I will also describe the distributed recommendation task which uses the three million Linux TiVo clients as well as the highly scalable, throttled, parallelized server-side architecture.

Joint work with Wijnand Van Stam.

My research interests lie at the boundary of application and theory in Information Extraction, Question Answering, Parse-based Feature classification, Bootstrap Learning, Sampling in databases, Active Learning and Bayesian Model Averaging.

My most recent set of papers is on sampling in databases based on an application fielded at Yahoo for over two years supporting thirteen internal analytics data-marts. Prior to that in the Web Search group, I did work on statistical evaluation and competitive analysis of search results which was important in Yahoo’s decision to acquire Inktomi. It also led to an ECIR paper on search evaluation framework.

My PhD is on Bayesian Model Averaging, which I received from UC Irvine. After that I did research and consulting at IBM Almaden and Stanford’s CLL lab before leaving academia for TiVo. At TiVo, I led the team that wrote the Suggestions Engine, a system for recommending TV shows which runs partly in distributed form on over three million Linux boxes (TiVo’s). Following TiVo, I was a principal scientist doing clickstream cluster analysis and text clustering at Vividence (now Keynote).