SOCIAL COMPUTING PROGRAM

Abstract

The social computing research program aims to develop models, algorithms, systems and tools to unravel the behaviorial knowledge about online user communities.  In this research, an integrated approach drawing expertise from both computing and social science disciplines will be adopted.   The program is currently funded by A*Star and is led by Ee-Peng Lim.  The other key members include Kuiyu Chang, Aixin Sun, and Anwitaman Datta.

Objectives

a. Design of a Semantic Social Network (SSNet) Framework to model and discover the behaviors of users and objects interacting with one another in an online community. Under this framework, we aims to define computational models for different important behaviors of people and objects based on interactions among them in an SSNet.

b. Development of efficient algorithms for realizing the computational models of behaviors. The efficacy and efficiency of these algorithms will be evaluated by experiments on social software data collected from Internet.

c. Design and implementation of a SSNet Engine. The SSNet engine aims to provide the storage, query and analytics services required by the existing social application software to exploit the discovered behaviors of people and objects to their advantages.

d. Application of SSNet framework to homeland security study. This research will apply SSNet to study the behaviors of Internet users and objects relevant to terrorism and other homeland security topics. Our focus will be on discovering defamatory/offensive objects and individuals/groups with prejudice behaviors, and accessing the impact of terrorism incidents on people. A behavior analysis application will be developed to track these behaviors in real-time and to alert domain experts the people/objects deserving attention.