A wealth of user activity happens on the Internet today, and understanding this web activity may lead to value in many dimensions. The Bing Predicts team (http://bit.ly/1wzumf5) is leveraging this data in a unique way to provide insightful predictions and projections to outcomes of events yet not decided. In the talk, we will discuss how machine learning techniques can be applied to a large set of aggregate data to parse out salient information to make accurate predictions on events in sports, politics, and entertainment.
Walter Sun received a B.S. degree in EECS from Georgia Tech, in 1995, and then obtained three graduate degrees at the Massachusetts Institute of Technology (M.S. in 1997, Engineer’s in 2004, Ph. D. in 2005) within the EECS department. Between his Master’s and Ph. D., Dr. Sun worked as a financial analyst at Blackrock, a software engineer at Apple Inc., & a Staff DSP engineer at 2Wire Inc. Since obtaining his Ph. D., he has worked the last 9 years at Microsoft Corporation, first in the Windows Codec team and now in the Bing.com organization. Currently, he works on Core Ranking of web results on Bing along with other machine learning-based projects like Bing Predicts. Dr. Sun is an adjunct faculty at Seattle University where he has taught courses in digital image processing, signals & systems, and digital logic.