This project aims to contribute to the systematic analysis and design of more reliable online reputation systems. This work is led by Chris Dellarocas and funded by the MIT eBusiness Center.
The production of trust is an important requirement for forming and growing open online trading communities. In such communities, the relative ease with which buyers and sellers can change partners, or even their identities, from one transaction to the next, gives incentives to both parties to provide inferior service quality or to hold back on their side of the exchange. Furthermore, many of the clues we draw on for establishing trust between partners in the real world a person's voice or body language, for instance are absent in cyberspace. These issues are even more pronounced in trading communities of software agents, where identity can be extremely volatile and common sense and intuition practically non-existent. Automated reputation reporting systems are emerging as an important trust management mechanism in online trading communities. The goal of reputation systems is to encourage trustworthiness in transactions by using past behavior as a publicly available predictor of likely future behavior. We are already seeing the first generation of such systems in the form of online ratings, feedback or recommender systems. As Web users grow to depend on them, online trust management systems deserve new scrutiny. Very little work has been published so far on the reliability of various online reputation systems and, even more importantly, on ways in which these systems can be compromised by colluding or unfair buyers and sellers.
Using a combination of theory, prototyping and simulation, the project objectives are to:
The Design of Reliable Trust Management Systems for Online Trading Communities. Working Paper.
Immunizing online reputation reporting systems against unfair ratings and discriminatory behavior. Proceedings of the 2nd ACM Conference on Electronic Commerce, Minneapolis, MN, October 17-20, 2000.