CATS-Oct-4-2013
Title
Algorithmic Game Theory of eBay's Buyer-Selling matching
Speaker
Dr. Kamal Jain is a scientist interested in the interface of technology and business innovation. He earned his doctorate from Georgia Tech in 2000 in areas consisting of computer science, mathematics, and operation research. He is well known in both academic community and technology industry. He has published 75+ research papers in top notch conferences and journals. One of his papers inspired a book on iterative methods in algorithms. He has also published 100+ patents, a third of which have already been granted. He enjoys mentoring. He has mentored dozens of PhD students, most of whom are now at top universities and companies.
Kamal, together with his wife and daughter, is a Seattle area resident since 2000. He spent 11+ years with Microsoft before he moved to eBay in 2011. At Microsoft he was involved in strategic research, such as search loyalty program and music business models. His timely research helped Microsoft to develop an alternate strategy than buying Yahoo entirely, thereby saving a fortune. His research work also inspired Microsoft's research center in Boston area. At eBay he tries to develop new insights on commerce from foundational perspective.
Kamal is also an Affiliated Professor of Computer Science at the University of Washington and a Charter Member of TIE at its Seattle chapter.
Abstract
At the broadest level, eBay is an intermediary, who is in the business of matching a buyer and a seller so that they can complete a mutually beneficial transaction. There are two important strategic choices eBay makes. First, how and whom does eBay charge a fee for its services. Two, how does eBay rank the possible product choices for a buyer.
In this presentation, we see eBay as a part of the intermediation
industry, e.g., Google is in the same business of intermediation where
they match buyer-seller (i.e., searcher-advertiser as known in their
world). We compare some of the popular intermediation business-models in
the industry. We then ask if the intermediary could choose any business
model, which would be a profit maximizing business model. We also try to
propose some algorithms to implement these business models.
Credit: This is a joint work with Chris Wilkens of Berkeley. The data analysis work is jointly done with Di Wang of Berkeley.