On June 7, 2011, Yahoo Labs will participate in the 12th ACM Conference on Electronic Commerce (EC’11), the leading scientific conference on advances in theory, systems, and applications for electronic commerce, hosted by the ACM Special Interest Group on Electronic Commerce (SIGECOM). Held in San Jose, EC'11 will discuss papers on a range of topics -- from applications, and empirical and experimental studies, to automation, personalization and targeting. Of the total accepted papers at EC’11, Yahoo Labs contributed six, focusing on hot topics such as managing user-generated content (UGC) and auctions. Of particular note is Arpita Ghosh’s involvement in the conference. Arpita, a senior research scientist at Yahoo Labs, contributed to three of the six accepted Yahoo papers. Her research generally focuses on algorithms and mechanism design in the context of the Web, particularly social computing and user-generated content, online advertising, and matching markets. Having joined Yahoo Labs in 2006 after finishing her PhD from Stanford, Arpita brings a wealth of expertise on her chosen subject areas. Brief descriptions of her accepted papers can be found below:

• A Game-Theoretic Analysis of Rank-Order Mechanisms for User-Generated Content (Ghosh and Hummel) – This paper analyzes rank-order mechanisms in UGC, proving that rank-order mechanisms, which display high-quality contributions above low-quality contributions, provide the right incentives for strategic contributors to generate high-quality content, and almost always incentivize higher quality contributions than more `fair’ proportional mechanisms.
• Who Moderates the Moderators? Crowdsourcing Abuse Detection in User-Generated Content (Ghosh, Kale and McAfee) – This paper comes up with a model for the problem of moderating UGC on popular online forums, and comes up with an algorithm that keeps spam and abusive content under control by smartly aggregating crowdsourced ratings from readers with unknown rating accuracy.
• Selling Privacy at Auction (Ghosh and Roth) – This paper proposes a model to embed privacy specifications into the sales of private data, showing how to reduce the problem of making accurate estimates from private data while compensating agents for their privacy loss from the computation to an auction design problem, designing truthful mechanisms to keep agents honest about their reported valuations for privacy.

Yahoo Labs, a sponsor of EC’11, will be presenting papers throughout the conference, which runs from June 7-9, 2011. Please see the full conference schedule here.