Case-based strategies for argumentation dialogues in agent societies

In multi-agent systems, agents perform complex tasks that require different levels of intelligence and give rise to interactions among them. From these interactions, conflicts of opinion can arise, especially when these systems become open, with heterogeneous agents dynamically entering or leaving the system. Therefore, agents willing to participate in this type of system will be required to include extra capabilities to explicitly represent and generate agreements on top of the simpler ability to interact. Furthermore, agents in multi-agent systems can form societies, which impose social dependencies on them. These dependencies have a decisive influence in the way agents interact and reach agreements. Argumentation provides a natural means of dealing with conflicts of interest and opinion. Agents can reach agreements by engaging in argumentation dialogues with their opponents in a discussion. In addition, agents can take advantage of previous argumentation experiences to follow dialogue strategies and persuade other agents to accept their opinions. Our insight is that case-based reasoning can be very useful to manage argumentation in open multi-agent systems and devise dialogue strategies based on previous argumentation experiences. To demonstrate the foundations of this suggestion, this paper presents the work that we have done to develop case-based dialogue strategies in agent societies. Thus, we propose a case-based argumentation framework for agent societies and define heuristic dialogue strategies based on it. The framework has been implemented and evaluated in a real customer support application.