Social Network Message Credibility: An Agent-Based Modelling Approach

The increased opportunities to engage in social interaction presented by Internet based platforms such as Facebook, My Space and Google+ has led to larger volumes of personal data being shared online. As such, social phenomena such as information sharing, propagation and ´gossiping´, amongst other behaviours, makes it challenging to access the integrity of online social interactions. Furthermore, the ability to remotely access, control and re-purpose text-based messages means that concerns about the security of this personal ´paper-trail´ of information is also becoming a higher priority for users of social media. Using an agent based modelling (ABM) approach, we use the abstraction of an ´agent´ to investigate how the dynamics of a social network can be affected by the presence of malicious agents. Such a scenario reduces the chances of having information that can be trusted as agents may be fabricating and sharing misleading information with other agents. However, the extent to which malicious agents can penetrate a network and the procedures by which credible data is achieved may not be known which is the primary thrust of this research.