News

Yahoo! Research Earns Highest Number of Accepted Papers at the Eighth ACM Conference on Electronic Commerce (EC '07)

The Yahoo! Research team was in fine form again, this time at the Eighth ACM Conference on Electronic Commerce (EC' 07) in sunny San Diego. Since 1999, the ACM Special Interest Group on Electronic Commerce (SIGECOM) has sponsored the leading scientific conference on advances in theory, systems, and applications for electronic commerce.

Yahoo! had the highest number of accepted papers of any institution – academic or industrial. The overall acceptance rate at the conference was 42/154 (27%). Of the accepted papers, eight had one or more Yahoo! co-authors, showcasing Yahoo’s dominant technical representation at the conference.

Several papers were presented by Yahoo! Scientists. Rica Gonen gave a talk on Generalized Trade Reduction, showing how to turn almost any auction-like mechanism into a budget balanced mechanism by reducing some trade. Sebastien Lahaie of Harvard, who will be joining Yahoo! Research in the Fall, presented a paper coauthored with David Pennock on improving revenue in search advertising auctions. Lahaie and Pennock analyze a family of auction rules that generalize the standard rank-by-bid and rank-by-revenue auctions used by search engines to sell advertising. Counterintuitively, ranking by "revenue" does not necessarily produce the maximum revenue for the search engine. Mohammad Mahdian co-authored two papers: one investigating how the compatibility of two technologies may effect the growth and spread of the newer technology in a social network, and one studying how to allocate online advertisement space with unreliable click or conversion estimates. Zoe Abrams, Ofer Mendelevitch, and John Tomlin authored a paper describing how to optimally deliver search advertisements when advertisers have budget caps. Michael Schwarz co-authored a paper studying bidding strategies in sponsored search auctions and Ravi Kumar and Yahoo! intern Esteban Arcaute co-authored a paper on query-incentive networks. Yiling Chen, David Pennock, and Evdokia Nikolova (a former Yahoo! intern) co-authored a paper about betting on permutations.

Other Yahoo! contributions included several well-attended tutorials and workshops. Michael Schwarz gave a tutorial with Kartik Hosanagar from the Wharton School of Business on Sponsored Search. The tutorial drew a crowd of about 50 people who were interested in learning about the multi-billion dollar Search Engine Marketing industry. Key topics that were covered included keyword generation, click fraud, click rate learning, contextual advertising, matching algorithms and targeting.

Yiling Chen gave a compelling tutorial entitled "Prediction Markets: Economics, Computation, and Mechanism Design." Attendees from a wide range of backgrounds were interested in economic topics such as how rational agents behave when facing risk and user interface models for a particular mechanism. They were also interested in recent research being conducted on prediction markets.

Yiling Chen and David Pennock co-organized a one day workshop on Prediction Markets, which drew a diverse crowd with backgrounds in economics, computer science, business, law, medicine and statistics. The program mixed theoretical, empirical and experimental studies and addressed the challenges and prospects for the future of prediction markets.

Two themes that dominated the conference were auctions and sponsored search. There was also a keen interest in social networks, embodied on the Web by the growing prominence of sites like Facebook, LinkedIn, and MySpace. Several workshops and tutorials focused on these major themes.

More info: http://stiet.si.umich.edu/ec07/