Arg2P: An argumentation framework for explainable intelligent systems


Roberta Calegari, Giuseppe Pisano, Andrea Omicini, Giovanni Sartor

Explainability of intelligent systems has become one of the main issues of today AI applications. Logic-based technologies, which characterised AI early days, might open new ways to engineer explainable intelligent systems, also for their long-term connection with multi- agent systems. In this paper we present the foundation of Arg2P, a logic-based argumentation framework for defeasible reasoning and agents’ conversation particularly suitable for reaching agents’ explainable intelligent behaviours. The Arg2P framework is presented and discussed both from a model and an architectural perspective to highlight its potential in intelligent systems engineering. Finally, a real application scenario is discussed in the domain of computable law.

(keywords) rg2P, intelligent systems engineering, explainable intelligent systems, logic- based technology, argumentation, defeasible reasoning, multi-agent systems

Journal of Logic and Computation,  2021, Oxford University Press.

Journals & Series

Tags: Arg2P

Publication

— authors

Roberta Calegari, Giuseppe Pisano, Andrea Omicini, Giovanni Sartor

— status

accepted with revision

Venue

— journal

Journal of Logic and Computation

— publication date

2021

BibTeX

— BibTeX category
article

Partita IVA: 01131710376 - Copyright © 2008-2021 APICe@DISI Research Group - PRIVACY