[DLS] Learning from Data: Popularity-Aware Topic Model
The ubiquitous use of mobile devices, combined with the emergence of cloud-based computing and storage infrastructure, has made it possible to trace every facet of human life. This data, when analyzed carefully, presents an unprecedented opportunity to learn about ourselves; how we behave, what we need, and what we love. The success of many online services, such as Facebook, Pandora, and Amazon, demonstrates how this data can be used to provide valuable services that users love. In this talk, I will explain a class of enormously successful learning models, probabilistic topic models, that have been used to learn about users and categorize their data for many online services.
Junghoo Cho is an associate professor in the Department of Computer Science at University of California, Los Angeles. He received a Ph.D. degree in Computer Science from Stanford University and a B.S. degree in physics from Seoul National University. His main research interests are in the study of the evolution, management, retrieval and mining of information on the World-Wide Web. He publishes research papers in major international journals and conference proceedings. He serves on program committees of top international conferences, including SIGMOD, VLDB and WWW. He is a recipient of the 10-Year Best-Paper Award at VLDB 2010, NSF CAREER Award, IBM Faculty Award, Okawa Research Award and Northrop Grunmann Excellence in Teaching Award.