Reasoning about Dynamic Preferences in Circumscriptive Theory by Logic Programming

To treat dynamic preferences correctly is inevitably required in the legal reasoning. In this paper, we present a method which enables us to handle some class of dynamic preferences in the framework of circumscription and to compute consistently its metalevel and object-level reasoning by expressing them in an extended logic program. This is achieved on the basis of policy axioms and priority axioms which permit to describe circumscription policy by axioms and play a role to intervene between the metalevel and object-level reasoning. Not only the information about preferences among rules and metarules but also relations between the dynamic preferences and priority axioms in circumscription are represented by a normal logic program. Thus priorities can be derived from the preferences dynamically, which allows us to compute the object-level circumscriptive theory by logic programming based on Wakaki and Satoh's method.