Prof. Devavrat Shah,
Jamieson Associate Professor,
Dept.of EECS, MIT
An important byproduct of the emergence of social networking platforms is an access to abundance of societal data in all forms: blogs, clicks, facebook feeds, transactions and tweets. It is of great interest to process this seemingly large volume of highly unstructured data to facilitate business decisions, public policy making or better social living. The key challenge lies in the fact that even though data is large in volume, the information content is very limited. Therefore, extracting meaningful answers has become a challenging computational and statistical task — the “Big Data” challange. In this talk, I will discuss how to resolve it successfully for important problems arising in the context crowd sourcing, group decision making, revenue management and viral advertising. The key to our success lies in the identification of the appropriate statistical framework for the problems at hand along with efficient message-passing algorithm.
Devavrat Shah is Jamieson Associate Professor with the department of EECS at MIT. He is affiliated with Laboratory for Information and Decision Systems and Operations Research Center. He completed his Ph.D. at Stanford University in October 2004. His thesis focused on the development of novel design and analytic methods for network algorithms. Before coming to LIDS in the fall of 2005, he spent a year at the Mathematical Sciences Research Institute (MSRI) in Berkeley, California. During this year of study, he was introduced to message-passing algorithms and graphical statistical inference. At LIDS, his research areas include statistical inference, network algorithms, and stochastics. He is the recipient of 2010 Erlang Prize from Applied Probability Society of INFORMS and 2008 ACM SIGMETRICS Rising Star Award. He currently serves as an associate editor of Operations Research.