计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (25): 210-211.DOI: 10.3778/j.issn.1002-8331.2008.25.063

• 工程与应用 • 上一篇    下一篇

基于蚁群算法信息素的异构案例集合构建策略

贾世杰,黄青松,刘利军   

  1. 昆明理工大学 信息工程与自动化学院,昆明 650051
  • 收稿日期:2008-03-03 修回日期:2008-06-10 出版日期:2008-09-01 发布日期:2008-09-01
  • 通讯作者: 贾世杰

Constructive strategy of non-isomorphic case set based on pheromone theory of ant colony algorithm

JIA Shi-jie,HUANG Qing-song,LIU Li-jun   

  1. College of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650051,China
  • Received:2008-03-03 Revised:2008-06-10 Online:2008-09-01 Published:2008-09-01
  • Contact: JIA Shi-jie

摘要: 基于CBR智能推荐系统是大型科学仪器协作共用网的重要组成部分。根据蚁群算法信息素更新原理设计并实现了一个完全异构案例集合构建策略。分析了完全异构案例集合构建原理,重点论述了案例权重动态分配的解决原理及过程。根据实验结果,表明该方法能够有效地提高智能推荐系统推荐结果的精确程度。

关键词: 基于案例推理, 特征项权重, 异构案例集合, 信息素

Abstract: Based on CBR,the intelligent recommending system is the important part of scientific instrument shared network.According to pheromone updating theory of ant colony algorithm,this paper designs a constructive strategy of complete non-isomorphic case set and carries it out.In this paper,the complete non-isomorphic case set theory is analyzed and the principle and process of how to solve the dynamic distribution of case weight is discussed.The results of a experiment show that this method can efficiently enhance the accuracy of the results of intelligent recommending system.

Key words: case-based reasoning, feature weight, non-isomorphic case set, pheromone