Multi-agent Information Diffusion Model for Twitter

During the 2011 East Japan Great Earthquake Disaster, many people used social media such as Twitter to get important information for their lives. But, generally, social media also has bad side, that is wrong information diffusion problem. In this paper, we will propose a novel multiagent-based information diffusion model, the Agent-based Information Diffusion Model (AIDM), and evaluate it. Up to now, our previous model is based on the SIR model, which is famous as a diffusion model of infection. The SIR model is represented by the stochastic state transition model for whether to propagate the information, and its transition probability is defined as the same value for all agents. However, people´s thinking or actions are not the same. To make person´s character heterogeneously, we adopted two elements in our proposal model: user diversity and multiplexing of information paths. From a comparison evaluation, it is shown that the proposed model basically to reproduce the information diffusion as same as the diffusion of real data.