Characterisation of a Novel Indexing Technique for Case-Based Reasoning

摘要

In this paper, we investigate two novel indexing schemes called D-HS and D-HS+PSR(II) designed for use in case-based reasoning systems. D-HS is based on a matrix of cases indexed by their discretised attribute values. D-HS+PSR(II) extends D-HS by combining the matrix with an additional tree-like indexing structure to allow for solution reuse. D-HS+PSR(II)’s novelty lies in its ability to improve retrieval efficiency over time by reusing previously encountered solution patterns. Benefits include simplicity, accuracy, speed, robustness to missing values and ability to facilitate efficient real time maintenance of retrieval knowledge as the size of the case-base grows. We present empirical results from an analyses of 20 case-bases and demonstrate the techniques to be of similar competency to C4.5 yet much more efficient. Performance advantages over C4.5 are shown to be especially apparent when tested on case-bases which grow in size over time or those with missing values.