We revisit the completion-based reasoning over incomplete Abstract Argumentation Frameworks (iAAFs), whose core is the notion of i-extension under the possible or necessary perspective, i.e. a set of arguments that, in some collective way, prove coherent and justified in at least one or in every completion, respectively. In particular, we show that, under the possible perspective, i-extensions exhibit a counterintuitive behavior, as they do not ensure conflict-freeness. Thus, we introduce the alternative notion of i*-extension, that not only fixes this behavior, but has also a positive computational impact: under various semantics of extensions, moving from i- to i*- extensions makes the complexity of the verification problem under the possible perspective move from NP-complete to P. Starting from this, we revisit the notion of accepted argument and define it as an argument belonging to at least one or every i*-extension, depending on whether the credulous or skeptical perspective is adopted. We show that this new definition does not in general coincide with the one in the literature (that is not based on the notion of i- or i*- extension), and can provide the analyst with a different perspective for reasoning over the justification of arguments in the presence of incompleteness. In this regard, we have thoroughly investigated the differences of the revisited acceptance problem with the acceptance problem in the literature, also in terms of computational complexity.

Revisiting the notions of extension and acceptance over incomplete abstract argumentation frameworks

Fazzinga B.;Flesca S.;Furfaro F.
2026-01-01

Abstract

We revisit the completion-based reasoning over incomplete Abstract Argumentation Frameworks (iAAFs), whose core is the notion of i-extension under the possible or necessary perspective, i.e. a set of arguments that, in some collective way, prove coherent and justified in at least one or in every completion, respectively. In particular, we show that, under the possible perspective, i-extensions exhibit a counterintuitive behavior, as they do not ensure conflict-freeness. Thus, we introduce the alternative notion of i*-extension, that not only fixes this behavior, but has also a positive computational impact: under various semantics of extensions, moving from i- to i*- extensions makes the complexity of the verification problem under the possible perspective move from NP-complete to P. Starting from this, we revisit the notion of accepted argument and define it as an argument belonging to at least one or every i*-extension, depending on whether the credulous or skeptical perspective is adopted. We show that this new definition does not in general coincide with the one in the literature (that is not based on the notion of i- or i*- extension), and can provide the analyst with a different perspective for reasoning over the justification of arguments in the presence of incompleteness. In this regard, we have thoroughly investigated the differences of the revisited acceptance problem with the acceptance problem in the literature, also in terms of computational complexity.
2026
Abstract argumentation
Complexity
Uncertain reasoning
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.11770/403257
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? 0
social impact