Home

Research

Publications

Teaching

Personal

James Caverlee :: Trusted Social Information Management

The past few years have seen the explosive rise of Web-based social networks, online social media sites, and large-scale information sharing communities -- all part of a Web 2.0 push that has attracted increasing media, industry, and research interest. Online communities are the fastest growing phenomenon on the Web, enabling millions of users to discover and explore community-based knowledge spaces and engage in new modes of social interaction. Web 2.0 successes include the social networking site MySpace, which has grown from 1 million user accounts in the first quarter of 2004 to an astonishing 250 million user accounts today; Wikipedia, with over 75,000 active contributors and more than 8 million user-generated articles; YouTube, which powers the viewing of over 100 million video clips per day; and on and on.

More generally, Web 2.0 can be viewed as an exemplar of Social Information Management. Social Information Management promises rich opportunities for information sharing, electronic commerce, and new modes of social interaction, all centered around the social Web of user-contributed content, social annotations, and person-to-person social connections. Social Information Management deviates from traditional information management through its emphasis on users as first-class objects in:

• the creation of data and knowledge artifacts;
• their active participation in the sorting, ranking, and annotation of other users and artifacts;
• the inherent social dynamics that impact social information management like group formation, member commitment, social connectivity, and social trust; and
• new modes of information discovery and knowledge management powered through social search and social navigation functionalities.

Understanding this new social phenomenon is a challenging research problem on its own. How can we maintain effective social connectivity and how do we build social trust in large-scale social networks? How do we structure social information spaces so that online communities can easily come together, attract new members and develop over time? What algorithms and architectures are needed to support social search, navigation, and knowledge discovery? Compounding these core questions is the challenge of enabling effective and efficient Social Information Management in the presence of malicious users in the social information space and the uncertainty associated with information and conclusions derived from such a space.

Existing Web 2.0 services have already observed many different vulnerabilities -- including impersonated (or fraudulent) digital identities, social network enhanced phishing, and corrupt user-generated metadata (or tags) -- and new threats are certain to emerge as attackers grow in sophistication. As more and more malicious entities target these systems, what guarantees can we make? Can we develop algorithms and architectures for the efficient, reliable, and secure use of social information spaces? Can we build systems in which users have trust in the data and information derived from these systems, even in the presence of users intent on undermining the quality of information?

In this research proposal, our goal is to develop and enable Trusted Social Information Management, both through the development of pro-active vulnerability analysis, detection, and elimination approaches and the development of trust-enhanced social information access algorithms.

As part of this research project, we have had some initial research successes. Please feel free to contact us for more information: