Séminaire d'équipe(s) Large-scale Heterogeneous DAta and Knowledge

Measuring Similarity between Logical Arguments
Victor David

06 March 2023, 00:00 Salle/Bat : 0/650
Contact :

Activités de recherche : Automated Reasoning

Résumé :

Argumentation is a prominent approach for reasoning with inconsistent information. It is based on the justification of formulas by arguments generated from propositional knowledge bases.
The aim of this talk is to present how to evaluate the similarity of two propositional logical arguments. For that purpose, we introduce a notion of similarity measure and a set of principles that such a measure should satisfy. We propose some intuitive extensions of measures from the literature and show that they fail to satisfy some of the principles.
Indeed, those measures may lead to inaccurate results when arguments are not concise, i.e., their supports contain information that is useless for inferring their conclusions. For circumventing this limitation, we start by refining arguments for making them concise. Then, we propose two families of similarity measures that extend existing ones and that deal with concise arguments.
Although these measures satisfy desirable properties, they suffer from the side effects of being syntax-dependent. Indeed, they may miss redundant information, leading to undervalued similarity. Finally, we overcomes this shortcoming by compiling arguments, which amounts to transforming their formulas into clauses,and using the latter for extending existing measures and principles. We show that the new measures deal properly with the critical cases.