|COOKIES: By using this website you agree that we can place Google Analytics Cookies on your device for performance monitoring.|
Learning item trees for collaborative filtering with implicit feedback
If you have a question about this talk, please contact Peter Orbanz.
User preferences can be inferred from either explicit feedback, such as item ratings, or implicit feedback, such as rental histories. Research in collaborative filtering concentrated on explicit feedback due to the ease of formalization, resulting in the development of accurate and scalable models. However, since explicit feedback is often difficult to collect, it is essential to develop effective models that take advantage of the more abundant implicit feedback.
We introduce a new approach to implicit feedback collaborative filtering based on modelling the item selection process performed by each user. In order to make it feasible to learn a different distribution over items for each user, we restrict our attention to tree-structured distributions. Since the accuracy of the resulting model is heavily dependent on the choice of the tree structure, we develop an algorithm for learning trees from data. Our algorithm is based on the online EM formalism and takes into account the probabilistic model the trees will be used with.
Joint work with Yee Whye Teh
This talk is part of the Machine Learning @ CUED series.
This talk is included in these lists:
Note that ex-directory lists are not shown.
Other listsHispanic Research Seminars CUSAS Forum Speaker Series Centre of African Studies Occasional Talks
Other talksThe Immunology of a successful pregnancy William Pitt Seminar and Dinner: The Anxiety of Age Signaling pathways controlling autophagy and its cross talk with apoptosis Oi! Scientology! Church Of Fear roadshow. Sex Education and Faith Communities in the UK: Challenges and Opportunities The influence of the Amundsen-Bellingshausen Seas Low on the climate of West Antarctica and its representation in coupled climate model simulations