东北大学学报（自然科学版）

东北大学学报(自然科学版) ›› 2006, Vol. 27 ›› Issue (1): 33-36.DOI: -

• 论著 • 上一篇    下一篇

基于事例推理系统中检索策略的分析与研究

黄玉基;魏伟杰;曾文;   

  1. 东北大学信息科学与工程学院;东北大学信息科学与工程学院;中国科学院沈阳自动化研究所 辽宁沈阳110004;辽宁沈阳110004;辽宁沈阳110016
  • 收稿日期:2013-06-23 修回日期:2013-06-23 出版日期:2006-01-15 发布日期:2013-06-23
  • 通讯作者: Huang, Y.-J.
  • 作者简介:-
  • 基金资助:
    国家“十五”科技攻关项目(2004BA721A05)

Analytical study on retrieving strategy in case-based reasoning system

Huang, Yu-Ji (1); Wei, Wei-Jie (1); Zeng, Wen (2)   

  1. (1) School of Information Science and Engineering, Northeastern University, Shenyang 110004, China; (2) Shenyang Institute of Automation, Chinese Academy of Science, Shenyang 110016, China
  • Received:2013-06-23 Revised:2013-06-23 Online:2006-01-15 Published:2013-06-23
  • Contact: Huang, Y.-J.
  • About author:-
  • Supported by:
    -

RichHTML

0

PDF (PC)

928

摘要： 讨论了基于事例推理(CBR)在人工智能系统中的作用,分析了常用的CBR检索算法,并着重研究了提高系统检索效率问题.针对CBR系统的关键性问题事例检索,提出了一种分级检索算法与最近相邻算法相结合的检索策略(L&NCBR),分析了在精确匹配与非精确匹配情况下该策略的效率,并给出了基于L&NCBR的智能推理系统运行情况分析.结果表明L&NCBR策略在提高系统检索的稳定性和效率方面是有效的.

关键词: 人工智能, CBR, 分级检索算法, 最近相邻算法

Abstract: Discusses the role of case-based reasoning (CBR) in artificial intelligence systems and some commonly used CBR algorithms. Then, the efficiency of CBR systems is emphatically studied. A new retrieval algorithm L and NCBR combining the nearest-neighbor algorithm with hierarchical retrieval algorithm is designed aiming at the crucial problem of case-based reasoning system. In addition, the efficiency of the algorithm under both accurate and inaccurate conditions is discussed as well as the running result of the L and NCBR-based intelligent system. The result showed that the strategy can improve the stability of the system and efficiency of retrieval.

中图分类号: