Information Diffusion in Online Social Networks: From Analyzing Trends to Stopping Misinformation

Social networks provide great opportunities for social
connection, learning, political and social change, as well as
individual entertainment and enhancement in a wide variety of forms.
Online social networks also provide unprecedented amounts of information
about social interaction and provide opportunities to study social
interactions on a scale and at a level of detail that has never been
possible before. In this talk, we will consider ways of systematically
exploring the vast space of on-line social network problems. Namely, we
will consider three dimensions; understanding, managing and reporting on
social networks and focus on example studies relating to these dimensions.
We will focus on two applications: limiting the spread of misinformation
in
social networks and trend analysis. In the former problem, we study the
notion of competing campaigns in a social network, and explore the spread
of influence in the presence of such competing campaigns. In particular,
we develop protocols whose goal is to limit the spread of misinformation
by identifying a subset of individuals that need to be convinced to adopt
the competing (or "good" ) campaign so as to minimize the number of people
that adopt the "bad" campaign . The second problem is motivated by the
critical role the identification of popular and important topics plays in
better understanding societal concerns. We propose two novel structural
trend definitions referred to as "coordinated" and "uncoordinated" trends
that use friendship information to identify topics that are discussed
among clustered and unconnected users respectively. Our analyses and
experiments show that structural trends are significantly different from
traditional trends and provide new insights into the way people share
information online.

This work was done jointly with Divyakant Agrawal and Ceren Budak.
My research interests are in the fields of fault-tolerant distributed systems and databases. I have designed several protocols for highly available fault-tolerant systems and have also been involved in designing systems and database support for geographic information systems and collaborative environments. Recently, I have been especially interested in exploring efficient ways for retrieving data in a distributed library as well as content-based searching using multi-dimensional index structures. I am currently involved in several research projects including Alexandria for building a distributed digital library (not to be confused with the ancient library of Alexandria ), an NSF sponsored data summarization project Yangtze , and an NSF sponsored project on leveraging advanced hardware for data stream managment