J4 ›› 2011, Vol. 49 ›› Issue (04): 733-739.

• 计算机科学 • 上一篇    下一篇

一种不均衡数据的改进蚁群分类算法

徐淑坦1, 王朝勇2, 孙延风1   

  1. 1. 吉林大学 计算机科学与技术学院, 长春 130012|2. 吉林工程技术师范学院 应用科学学院, 长春 130052
  • 收稿日期:2010-07-26 出版日期:2011-07-26 发布日期:2011-08-16
  • 通讯作者: 孙延风 E-mail:sunyf@jlu.edu.cn

An Improved AntMiner Algorithm for Unbalanced Data

XU Shutan1, WANG Chaoyong2, SUN Yanfeng1   

  1. 1. College of Computer Science and Technology, Jilin University, Changchun 130012, China;2. School of Applied Science, Jilin Teachers Institute of Engineering and Technology, Changchun 130052, China
  • Received:2010-07-26 Online:2011-07-26 Published:2011-08-16
  • Contact: SUN Yanfeng E-mail:sunyf@jlu.edu.cn

摘要:

针对蚁群挖掘算法(ant colony mining algorithm, ACMA)中的规则评价函数和规则修剪方法, 提出一种改进的蚁群挖掘算法(improved ant colony mining algorithm, IACMA), 并将其应用于不均衡数据分类. 数值实验采用基准数据库中3种典型的不均衡数据, 结果表明, 改进后的算法能有效提取少数类, 提高了不均衡数据整体分类效果.

关键词: 不均衡数据分类; 蚁群分类算法; 蚁群挖掘算法; 数据挖掘; 规则提取

Abstract:

Based on the quality function and pruning method of ant colony mining algorithm (ACMA), an improved ant colony mining algorithm (IACMA)
 was proposed and applied to unbalanced data classification. Three datasets from the typical benchmark database were used for the numerical experiment. The simulation results show that the IACMA can better process the minor categories, and improve the overall classification accuracy.

Key words: unbalanced data classification, AntMiner, ant colony mining algorithm, data mining, rule extraction

中图分类号: 

  • TP18