Learning and Inference of Large-Scale Event Data¶
Hongyuan Zha
Professor
Georgia Institute of Technology
Abstract: Dynamic processes, such as rumor spreading in social networks, occurrence of crimes in a city, migration of birds across continents, generate a large volume of high dimensional "asynchronous" and "interdependent" temporally and spatially stamped event data. This type of event data is rather different from traditional iid. data and discrete-time temporal data, which calls for new models and scalable algorithms for analyzing, learning and utillizing them. In this talk, I will present a framework based on multivariate point processes, high dimensional sparse recovery, and randomized algorithms for addressing a sequence of problems arising from this context. As a concrete example, I will also present experimental results on leanring and optimizing information diffusion in web logs, including estimating hidden diffusion networks and influence maximization with the learned networks. With both careful model and algorithm design, the framework is able to handle millions of events and millions of networked entities, and achieve the staate-of-the-art results.
Bio: Hongyuan Zha is a Professor at Software Engineering Institute, ECNU and the School of Computational Science and Egineering, College of Computing, Georgia Institute of Technology. He earned his PhD degree in scientific computing from Stanford University in 1993. Since then he has been working on information retrieval, machine learning applications, and numerical algorithms. Before joining Georgia Tech, Hongyuan Zha was a Professor at the Department of Computer Science and Engineering at Pennsylvania State University from 1992 to 2006, and worked from 1999 to 2001 at Inktomi Corporation.