A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Information and Communications Technologies, School of Engineering and Technology

Thesis (Ph.D.) - Asian Institute of Technology, 2013

Over the years, many distinct and well-known approaches to reasoning with inconsistent and incomplete knowledge bases (like logic programming, Reiter's default logic, Poole's Theorist, coherence-based systems, autoepistemic logics and argumentation among others) have been developed and studied extensively. Given the diversity of the motivation and , backgrounds of these distinct approaches, it is unlikely that a single framework will ever be adopted for all applications even though argument-based systems are becoming increasingly popular due to their intuitive appeal to the ways humans perform their practical and daily reasoning. An appealing idea to deal with the diversity is to integrate the systems in' such a way to allow the reasoners to use whatever approach fits their purposes best. We argue that a simple way to integrate argument-based and non-argument-based systems is by 1) translating non-argument-based systems into argument-based ones and 2) proposing a common methodology for stepwise development of proof procedures for argument-based systems. In this way, reasoners could be offered a "front-end" encompassing whatever sys- tems they find appealing, and at the same time, new proof procedures need to be developed for argument-based systems only. This simple integration concept would reduce the need to develop proof theories and procedures for many diverse systems at the "front-end" and take advantage of the proof procedures and theories developed for argument-based systems like assumption-based argumentation or logic programming with industry-strength technology like Prolog. It has been showed that logic programming, Reiter's default logic, Poole's theo- rist or autoepistemic logics could be captured naturally by assumption-based argumentation though it is still an open question whether coherence systems could be captured by argu- mentation at all. Much research has also been done to study dialectical proof procedures for argumentation in the literature, either for abstract argumentation, or for rule-based instances of it like logic programming or assumption-based argumentation though there is no common methodology for their development and hence, every time, when there is a need for a new argument-based proof procedure, such a procedure as well as its soundness and complete- ness will have to be developed and studied from scratch without benefiting much from the insights and results from the development of previous ones. This thesis consists of two parts. In the first part, we show that Brewka's preferred sub- theories, arguably the most influential among coherence systems, could be translated effec- tively and naturally into assumption-based argumentation. In the second part, we propose a systematic stepwise methodology for development of dialectical proof procedures for argu- mentation. Our new results could be viewed as a relevant step towards a "proof-of-concept" of the simple integration of diverse systems for reasoning with incomplete and inconsistent knowledge bases.