Refined Preference-based Argumentation Frameworks
Kaci Souhila
Abstract
Kaci Souhila
Abstract
Argumentation is a reasoning model based on constructing arguments, determining potential conflicts between arguments and determining acceptable arguments. Dung's argumentation theory is an abstract framework based on a binary defeat relation between arguments. Due to this abstract representation, it has been instantiated in different ways. In particular, preference-based argumentation frameworks take into account a preference relation over arguments together with a (non necessarily symmetric) attack relation. We show that preference-based argumentation frameworks faithfully instantiate Dung's framework only when the attack relation is symmetric. Moreover the latter condition prevents undesirable results. We also promote a higher impact of preferences in preference-based argumentation frameworks and propose different ways to rank-order sets of acceptable arguments.
OpenAlex reports 14 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Argumentation is a reasoning model based on constructing arguments, determining potential conflicts between arguments and determining acceptable arguments. Dung's argumentation theory is an abstract framework based on a binary defeat relation between arguments. Due to this abstract representation, it has been instantiated in different ways. In particular, preference-based argumentation frameworks take into account a preference relation over arguments together with a (non necessarily symmetric) attack relation. We show that preference-based argumentation frameworks faithfully instantiate Dung's framework only when the attack relation is symmetric. Moreover the latter condition prevents undesirable results. We also promote a higher impact of preferences in preference-based argumentation frameworks and propose different ways to rank-order sets of acceptable arguments.
Key concepts: Argumentation theory, Preference, Computer science, Psychology, Epistemology, Mathematics, Philosophy, Statistics