Case-Base Maintenance Based on Multi-Layer Alternative-Covering Algorithm

Case-based reasoning systems running in interactive domains like e-commerce, can easily reach thousands of cases which are stored in the irreducible case library, and the deletion of any single case means that the uniquity is lost. Aiming on the efficiency of retrieval in this kind of environment, this paper proposes the methods to achieve case-base maintenance (CBM) from the both sides: one is employing alternative-covering algorithm to partition the case library to many covering domains and thus realizing the selective filtering; the other is using multi-layer feedforward neural networks to deal with case retrieval within the large-scale case library. Our experimental results indicate that the integrated method, which is especially feasible for the processing of vast and high dimensional data, can effectively guarantee the system's usability and enhance its capability