关键词: 基于案例推理, 聚类, 目标案例, 相似度

Abstract: Aiming at the problems concerning case retrieval algorithm in the CBR system，this paper，in the light of the idea of the k-means algorithm，firstly clusteres the case database and then works out a case retrieval algorithm on the basis of the clustering of the case database.It analyzes the selecting principles of sample case，and mainly discusses the case retrieval algorithm.The results of experiment show that this algorithm can efficiently enhance the recall rate of the case retrieval outcomes and the efficiency of the case retrieval.

Key words: case-based reasoning, clustering, target case, similarity