系统工程与电子技术

• 软件、算法与仿真 • 上一篇    下一篇

基于首次找准原则和相邻原则的LF蚁群聚类的改进算法

薛松1,3, 陈寿文4, 丰景春1,2   

  1. 1. 河海大学商学院, 江苏 南京 211100; 2. 河海大学项目管理研究所, 江苏 南京 211100;
    3. 江苏省“世界水谷”与水生态文明协同创新中心, 江苏 南京 211100;
    4. 滁州学院数学与金融学院, 安徽 滁州 239000
  • 出版日期:2017-02-25 发布日期:2010-01-03

LF improved colony clustering algorithm based on capturing for first time and proximity principles

XUE Song1,3, CHEN Shouwen4, FENG Jingchun2   

  1. 1. Business School, Hohai University, Nanjing 211100, China; 2. Institute of Project Management, 
    Hohai University, Nanjing 211100, China; 3. Jiangsu Provincial Collaborative Innovation Center of 
    World Water Valley and Water Ecological Civilization, Nanjing 211100, China; 4. School of 
    Mathematics and Finance, Chuzhou University, Chuzhou 239000, China
  • Online:2017-02-25 Published:2010-01-03

摘要:

研究了LF蚁群聚类算法的改进问题。首次找准原则能够克服算法运行初期蚂蚁长时间遇不到待处理对象的现象,使得不同的蚂蚁一开始就能遇上不同对象。在LF算法中,当蚂蚁未负载且对当前对象试图执行拾起操作时,若不转移该对象,则采用位置相邻原则来预选下一个待处理对象。当蚂蚁负载且对当前对象试图执行放下操作时,则采用对象间相邻原则来选择相应的放置位置。改进后的算法既可以独立处理聚类问题,又可以作为其他算法的预处理步骤,具有一定的应用价值。

Abstract:

An LF improved colony clustering algorithm is discussed. Capturing for first time principle can overcome the phenomenon that ants can not find the object to be processed in the initial period of algorithm operation, so that different ants can meet different objects at the beginning. In the LF algorithm, when ants are not loaded and try to execute pick up operation instructions, if they do not transfer the object, the next object to be processed will be chosen based on the location proximity principle. When ants are loaded and try to execute put down operation instructions, the appropriate placement will be selected based on the proximity principle of objects. The improved algorithm can not only deal with the clustering problem, but also can be used as a preprocessing step for other algorithms, and it has a certain application value.