Computing the grounded semantics in all the subgraphs of an argumentation framework: An empirical evaluation

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Given an argumentation framework AF = (Ar, R) - with Ar a finite set of arguments and R ⊆ Ar x Ar the attack relation identifying the graph G - we study how the grounded labelling of a generic argument a ∈ Ar varies in all the subgraphs of G. Since this is an intractable problem of above-polynomial complexity, we present two non-naïve algorithms to find the set of all the subgraphs where the grounded semantic assigns to argument a specific label l ∈ {in, out, undec}. We report the results of a series of empirical tests over graphs of increasing complexity. The value of researching the above problem is two-fold. First, knowing how an argument behaves in all the subgraphs represents strategic information for arguing agents. Second, the algorithms can be applied to the computation of the recently introduced probabilistic argumentation frameworks.

Original languageEnglish
Title of host publicationComputational Logic in Multi-Agent Systems - 14th International Workshop, CLIMA 2013, Proceedings
Pages119-137
Number of pages19
DOIs
Publication statusPublished - 2013
Event14th International Workshop on Computational Logic in Multi-Agent Systems, CLIMA 2013 - Corunna, Spain
Duration: 16 Sep 201318 Sep 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8143 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference14th International Workshop on Computational Logic in Multi-Agent Systems, CLIMA 2013
Country/TerritorySpain
CityCorunna
Period16/09/1318/09/13

Keywords

  • Algorithms
  • Argumentation Theory
  • Semantics

Fingerprint

Dive into the research topics of 'Computing the grounded semantics in all the subgraphs of an argumentation framework: An empirical evaluation'. Together they form a unique fingerprint.

Cite this