An Argumentation-Based Interaction Model and its Algorithms in Multi-agent System

Argumentation-based interaction model in multi-agent system is based on the idea that agents not only send messages, but also give reasons to support those messages. In this paper, we propose an argumentation-based interaction model between agents and algorithms for argument evaluation. In the model, a message is constructed as the form of argument which has one or more premises and only one conclusion like a non-monotonic rule in argumentation-based formal system. The process of interaction between agents is constructed as an abstract dialogue tree with nodes representing arguments, and edges describing relations between arguments. The proposed algorithms are used to determine the defensibility of arguments and the acceptability of statements. Finally, an example is given to verify the validity of the model.