Recommendation with an Understandable Reason: Collaborative Topic Modeling...

Recommendation with an Understandable Reason: Collaborative Topic Modeling for Recommending Scientific Articles.

We're excited to have Chong Wang from Princeton at the helm in December. His talk description and bio follow:

Description:

Recommender systems are important and ubiquitous. Lots of work has been done to improve their prediction abilities, for example, the famous netflix competition. However, most of these systems seem to lack easily interpretable structures, which could be very important for end user interactions and system diagnosis. In this talk, I will talk about a new model that mitigates this issue on the application of recommending scientific articles. I will first review the matrix factorization for collaborative filtering and probabilistic topic modeling, then present our new algorithm of collaborative topic modeling for recommending scientific articles, which combines the merits of traditional collaborative filtering and probabilistic topic modeling. Our model provides an interpretable latent structure for users and items, and can form recommendations about both existing and newly published articles. We study a large subset of data from CiteULike, a bibliography sharing service, and show that our algorithm provides a more effective recommender system than traditional collaborative filtering. This is joint work with David Blei.

Bio:

Chong Wang is currently pursuing his Ph.D. in the Computer Science Department at Princeton University, under the supervision of Prof. David M. Blei. He obtained his B.E. and M.E. from Tsinghua University, both with honors. Then, he worked as an assistant researcher in the Web Search and Data Mining group, Microsoft Research Asia. Since arriving at Princeton, he has published many papers in leading international conferences and won several awards (best student paper award at KDD 2011 and notable paper award at AISTATS 2011). His ultimate goal is to make machine learning results more friendly to end users. He received the Google Ph.D. Fellowship for machine learning in 2010. He is also a Siebel Scholar, class of 2012.