Reasoning about Trust and Time in a System of Agents

The study of trust in Multi-Agent Systems (MASs) has been an area of interest for many researchers over the last years. This is due to the fact that trust is the basis for agent communication wherein entities have to operate in a dynamic and uncertain environment. Several approaches have been proposed to define logical semantics for trust in MASs. However, these approaches are limited to reason about trust based on the sole agents' mental states. Therefore, this paper considers trust from a high-level abstraction based on the social correct behaviors of agents. Specifically, we propose a logical framework that allows us to reason about unconditional trust and time. In particular, we introduce a new logical language called Trust Computation Tree logic (TCTL) that extends the Computation Tree Logic (CTL) with a new modality to represent trust. We describe the semantics by extending the interpreted systems formalism and consider a set of reasoning rules along with proofs to support our logic. Finally, we evaluate our approach using a real-life case study in the e-business domain to explain our proposed logic in a practical application.