collaborative topic modeling for recommending scientific.pdfVIP

Collaborative Topic Modeling for Recommending Scientific Articles Chong Wang David M. Blei Computer Science Department Computer Science Department Princeton University Princeton University Princeton, NJ, 08540, USA Princeton, NJ, 08540, USA chongw@ blei@ ABSTRACT effective practice—and one that we should continue—but it limits Researchers have access to large online archives of scientific arti- researchers to specific citation communities, and it is biased towards cles. As a consequence, finding relevant papers has become more heavily cited papers. A statistician may miss a relevant paper in difficult. Newly formed online communities of researchers sharing economics or biology because the two literatures rarely cite each citations provides a new way to solve this problem. In this paper, we other; and she may miss a relevant paper in statistics because it was develop an algorithm to recommend scientific articles to users of an also missed by the authors of the papers that she has read. One of online community. Our approach combines the merits of traditional the opportunities of online archives is to inform researchers about collaborative filtering and probabilistic topic modeling. It provides literature that they might not be aware of. an interpretable latent structure for users and items, and can form A complementary