Abstract: | We present a case-based reasoning system which can assist domain experts in interpreting building regulations in the relaxation process. When a relaxation application is submitted, experts make a decision on relaxation through consultation, in which relevant information is gathered and a variety of assertations by the parties concerned are put forward and considered by those experts. The aim of our system is to build knowledge bases of regulatory information and case histories and to provide experts with relevant information so that they can make an informed decision on relaxation. In order to provide a framework for the interpretation of building regulations, the system maintains abstraction hierarchies of legal rules from the statutory regulations and previously relaxed cases. The system retrieves legal rules and cases relevant to the input case by using the notion of implied similarity as well as structural mapping. It also records argumentation in the case library to explain key points when a similar case arises subsequently. The system has been implemented and tested with 22 cases derived from actual relaxation decisions.
|