An Argumentation Framework for Decision Making

Maxime Morge

Since a decision can be resolved by confronting and evaluating the justifications of different positions, argumentation can support such a process. This is the reason why many works in the area of Artificial Intelligence focus on computational models of argumentation. In particular, nonmonotonic logic techniques have been used as a model with hierarchies of possibly conflicting rules. However, even if modern techniques are used, this logical approach is still limited to the epistemic reasoning. The point is that a decision is not limited to draw conclusions. For this purpose, we propose an argumentation framework for practical reasoning. A logic language is used as a concrete data structure for holding the statements like knowledge, goals, actions. Different priorities are attached to these items corresponding to the user's preferences, the likelihood of the knowledge and the credibility of alternatives. These concrete data structures consist of information providing the backbone of arguments. Due to the abductive nature of decision making, arguments are built by reasoning backwards. Moreover, arguments are defined as tree-like structures. In this way, our framework suggests some solutions and provides an intelligible explanation of this choice.

This work is supported by the Sixth Framework IST programme of the EC, under the 035200 ARGUGRID project.

About the speaker: http://www.di.unipi.it/~morge/