In this talk, David Blei will review the basics of topic modeling and describe his recent research on collaborative topic models—models that simultaneously analyze documents and the corresponding reader behavior. Blei will explain how using collaborative topic models to discover patterns in how people read can help point readers to relevant new documents. Finally, he will discuss the broader field of probabilistic modeling, which gives data scientists both a rich language for expressing statistical assumptions and scalable algorithms for using those assumptions to uncover hidden patterns in massive data.
David Blei is associate professor of computer science at Princeton University. Blei’s research focuses on developing methods for finding patterns in large datasets. He has received awards including a Sloan Fellowship (2010), the Office of Naval Research Young Investigator Award (2011), the Presidential Early Career Award for Scientists and Engineers (2011), the Blavatnik Faculty Award (2013) and the ACM-Infosys Foundation Award (2013). Blei earned his Ph.D. in computer science from the University of California, Berkeley.
If this lecture is videotaped, it will be posted here after production.