2:00 – 2:05 — Introduction

2:05 – 3:00 — Keynote speech: Exploring User Opinions in Recommender Systems, Bing Liu

3:00 – 3:20 — A Modified Fuzzy C-Means Algorithm For Collaborative Filtering, Jinlong Wu and Tiejun Li

3:20 – 3:40 — Putting the collaborator back into collaborative filtering, Gavin Potter

3:40 – 4:00 — break

4:00 – 4:30 — From hits to niches? or how popular artists can bias music recommendations, Oscar Celma and Pedro Cano

4:30 – 5:00 — Investigation of Various Matrix Factorization Methods for Large Recommender Systems,Domonkos Tikk, Gabor Takacs, Istvan Pilaszy and Bottyan Nemeth

5:00 – 5:30 — Improved Neighborhood-Based Algorithms for Large-Scale Recommender Systems, Andreas Toescher, Michael Jahrer and Robert Legenstein

Keynote speech:
Exploring User Opinions in Recommender Systems
Bing Liu

Abstract: Traditionally, recommender systems operate based on user-behavior and rating data at the personal and/or aggregate level. In this talk, I will try to go beyond this tradition to discuss some new/future developments of recommender systems, even general advertising systems for that matter, based on opinions on the Web (e.g., in reviews, forum discussions, blogs, etc). Recommendations based on such data can be highly targeted and can also be embedded widely in the most appropriate context. Needless to say, I will introduce some recent developments in the area of opinion mining and sentiment analysis, and discuss whether these developments are ready for prime time.