Computationally-Friendly Argumentation Schemes

Abstract

In an agent activity both paradigmatic aspects of communication and reasoning can be naturally enriched by argumentation. The contribution of this research is a computationallyfriendly framework for formalizing paraconsistent argumentation schemes in information-rich environments.This goal is achieved by extending agent's reasoning capabilities with non-deductive methods like argumentation skills. We provide a generic paraconsistent program template permitting for implementation of various argumentation schemes. To do this, we appeal to techniques specific for the four-valued framework of 4QL, the rule-based, DATALOG ¬¬ -like query language. Although dealing with paraconsistency in argumentation schemes is not a new subject, our computational approach takes advantage of the tractability of 4QL. The paper concludes with an example of the Expert Opinion scheme implemented in 4QL. I. MODELING ASSUMPTIONS A. Realistic Models of AgencyComplex interactions is an immanent property of multiagent systems (MAS). They are vital to all paradigmatic activities like coordination, collaboration and negotiation, which naturally include phases of communication and reasoning. Communication has a long tradition as an important topic in computer science, specifically in (distributed) artificial intelligence and recently in MAS [6], [21]. Starting from fixed communication protocols in distributed systems, we now attempt to approach flexible dialogues among agents (see e.g., [5]).In an agent activity both aspects of communication and reasoning can be naturally enriched by argumentation. Therefore in the current paper we introduce a framework to formalize argumentation schemes. In order to address problems appearing in realistic domains we consider them in the presence of uncertain and inconsistent information, assuming that four types of situations may occur:• fact a holds,• fact a does not hold,• it is not known whether a holds, • information about a is inconsistent. Following Dunin-Kȩplicz and Szałas [11], the way the individual agents deal with conflicting or lacking information is encoded in their epistemic profile (see Section III). In typical multi-agent settings epistemic profile embodies agents' reasoning capabilities. These in turn influence the agent's deductive processes to finally affect their belief structures, i.e., agents' informational stance [11]. Moreover, various agents may reason differently using different methods of information disambiguation. In this research we will show how to augment an agent epistemic profile by such non-deductive reasoning capabilities like argumentation schemes. Supported by the Polish National Science Centre grants 2011/01/B/ST6/02769 and CORE 6505/B/T02/2011/40 B. Why 4QL: Realistic Representation and TractabilityOur approach is strongly influenced by ideas underlying 4QL: a four-valued paraconsistent query language introduced by Małuszyński and Szałas (see [16], [17] and Section V for details) 1 . 4QL supports a modular and layered architecture, providing simple, yet powerful constructs...