Reasoning with imperfect context and preference information in multi-context systems

Abstract

Multi-Context Systems (MCS) are logical formalizations of distributed context theories connected through a set of mapping rules, which enable information flow between different contexts. Recent studies have proposed adding non-monotonic features to MCS to handle problems such as incomplete, uncertain or ambiguous context information. In previous work, we proposed a non-monotonic extension to MCS and an argument-based reasoning model that enable handling cases of imperfect context information based on defeasible reasoning. To deal with ambiguities that may arise from the interaction of context theories through mappings, we used a preference relation, which is represented as a total ordering on the system contexts. Here, we extend this approach to additionally deal with incomplete preference information. To enable this, we replace total preference ordering with partial ordering, and modify our argumentation framework and the distributed algorithms that we previously proposed to meet the new requirements. © 2010 Springer-Verlag.