Reasoning, Arguing, Ranking and Aggregating
This work is composed of three main parts. The first part is devoted to the semantics of argumentation and follows an approach based on axiomatic principles. We study extensional semantics and graded semantics. We also identify the links between argumentation-based reasoning and reasoning based on other systems, for example that based on maximal coherent subsets of a knowledge base. We also study the links between extensional semantics and graded semantics and combine the two.
The second part studies how to reduce the number of extensions in order to obtain more conclusions. When the number of extensions is high, there may be no, or very few, conclusions. Since conclusions are obtained from the intersection of extensions, reducing the number of extensions can be beneficial in terms of inferential power.
The third part addresses the problems associated with aggregating a finite set of individual judgments, based on responses to a set of logically related questions. Although each individual has a set of coherent judgments, the question is how to aggregate them in order to obtain a set of coherent collective judgments, which should represent the opinions of individuals. The goal of my research in judgment aggregation is to study the properties of operators, as they provide a better understanding of the operators and the links between them.
- Pietro Baroni, Università degli Studi di Brescia
- Anthony Hunter, University College London
- Francesca Toni, Imperial College London
- Sébastien Konieczny, CNRS, CRIL, Univ. Artois
- Pierre Marquis, Univ. Artois, CRIL, CNRS
- Nicolas Maudet, Sorbonne Université - LIP6