Rapid Retrieval Algorithms for Case-Based Reasoning

Abstract

One of the major issues confronting case-based reasoning (CBR) is rapid retrieval of similar cases from a large case base. This paper describes three al gorithms which address this problem. The first algorithm works with quantitative cases using a graphical paradigm where the hyperspace containing th e cases is divided into smaller and smaller hypercubes. The retrieval time for this algorithm is O(Log(N)), where N is the number of cases. The second algorithm works on qualitative data by efficiently ·retrieving cases based on every necessary combination of case attributes. Its retrieval time varies only with respect to the number of attributes. The third algorithm is a combination of the previous two and allows retrieval of cases consisting of both quantitative and qualitative information. The al gorithms described in this paper are the first practical algorithms designed for case based retrieval on very large numbers of cases. The algorithms easily handle case bases containing millions of cases or more.