AMAM: Adaptive Multi-Agents Based Model for Negative Key Players Identification in Social Networks

Abstract

Social Network Analysis (SNA) is an active research topic. It arises in a broad range of fields. One important issue in SNA is the discovery of key players who are the most influential actors in a social network. Negative Key Player Problem (KPP-NEG) aims at finding the set of actors whose removal will break the social network into fragments. By another way, Multi-Agents Systems (MAS) paradigm suggests suitable ways to design adaptive systems that exhibit desirable properties such as reaction, learning, reasoning and evolution. A fortiori, the intrinsic nature of social networks and the requirements of their analysis could be efficiently handled using a MAS framework. Within this context, this paper proposes a multi-agents based-model AMAM for KPP-NEG. We first represent the social network in terms of a weighted graph. Then, a set of agents cooperate in order to identify the most important nodes. Simulation and computational results are demonstrated to confirm the effectiveness of our approach.