摘要
如何有效地建立简洁■性能可靠的范例库,以提高基于范例推理系统的性能,是当前基于范例推理研究的热点.本文结合NCL CLARA聚类算法与脚标数据的优点.给出了一种有效的基于能力的范例库自动建立新方法.通过实验表明,该方法能在保持原始领域数据库系统解决问题能力的前提下,最大程度地减少生成的范例库.
An important focus of recent CBR research is on how to develop strategies for achieving compact, competent case-bases, as a way to improve the performance of CBR systems. In this paper, a new method of Case-Base Automatic Setup is proposed. This method absorbs merits of NCL CLARA clustering algorithm and footprint data. The experiments show that the new algorithm can reduce cases greatly in Case-Base and preserve competence of original domain database system.
出处
《模式识别与人工智能》
EI
CSCD
北大核心
2004年第3期306-310,共5页
Pattern Recognition and Artificial Intelligence
基金
国家自然科学基金(No.70171052,90104030)
皖泰开发基金(No.143-150401)
关键词
基于范例推理
能力
脚标
聚类分忻
Case-Based Reasoning
Competence
Footprint
Clustering
