Man is many things, but he is not rational.

- Oscar Wilde

Social influence analysis in online social networks is the study of people’s influence by analyzing the social interactions between individuals. Since the last decade, there have been numerous research efforts to understand the influence propagation phenomenon due to its importance to viral marketing and information dissemination among others. Despite the progress achieved by state-of-the-art social influence analysis techniques, a key limitation of majority of these techniques is that they transform the problem to a cold hard network algorithm problem by stripping off fundamental factors of human psychology and behaviors. For example, these efforts ignore human psychology such as conformity of people, which refers to a person’s inclination to be influenced. Consequently, despite the great progress made in terms of algorithmic efficiency and scalability, existing techniques may not necessarily produce high quality results in practice. In this research, we investigate the interplay between psychology and social influence in online social networks and devise novel social influence-based solutions that are psychology-aware. In particular, our research aims to bridge two disparate fields, social psychology with online social analytics, and address some interesting problems at their intersection.

Our research results have appeared in premium venues such as ACM SIGMOD and VLDB Journal.

This research is partially funded by ROAR grant from NTU.

Key Achievements

  • CASINO is the world's first framework that incorporates a key human pyschology, conformity, in computing social influence (ACM CIKM 2011).
  • CINEMA is the world's first influence maximization solution that considers conformity of users during influence propagation (VLDB J, EDBT 2013).
  • GetReal is the first framework that takes a major step forward towards making competitive influence maximization realistic by considering social behavior of competitors (ACM SIGMOD 2015).
  • CHASSIS is the world's first framework that incorporates conformity in Hawkes Processes-based online information diffusion model (ACM SIGMOD 2020).


The list of publications related to this project can be found in ResearchGate.