Trust Management in Open Information Repositories

Students: Sara Javanmardi, UC Irvine/ISR

Advisor: Cristina Videira Lopes, UC Irvine/ISR

Abstract: Open information infrastructures available on the current Web allow millions of users to share information through a growing collection of tools and platforms. People can publish content and collaborate using social software infrastructures such as wikis, blogs, podcasts and shared forums. However, the open nature of such infrastructures makes it difficult for users to determine the reliability of the available information. For example, the trustworthiness of Wikipedia, one of the most visible public information repositories, is sometimes put into question, because it allows anyone to edit most of its content without ensuring that those editors are well-intentioned or well-qualified. In this paper, we present a trust model for open information repositories. The trust model estimates the reliability of the content based on the reputation of the contributors. Content reliability and contributors' reputations are updated dynamically, according to the users' actions in the system. This dynamics is modeled as a Markov process. As empirical validation of the concept, we implemented a trust layer that was emulated using actual usage patterns of Wikipedia. The results demonstrate the validity and usability of our trust management system.

Bio:

Sara Javanmardi is a graduate student in the Department of Informatics at University of California, Irvine. Her research interests include Information Security in Ubiquitous Computing and Semantic Web environments.