基于蚁群优化的分类规则挖掘方法  

Mining Classification-Rule Method Based on Ant Colony Optimization Algorithm

作　　者：束建华[1] 倪志伟[2] 杨善林[2] 

机构地区：[1]合肥工业大学计算机与信息学院 [2]合肥工业大学管理学院,安徽合肥230009

出　　处：《广西师范大学学报（自然科学版）》2007年第4期18-23,共6页Journal of Guangxi Normal University:Natural Science Edition

基　　金：国家自然科学基金子基金重点项目基金资助(70631003);安徽省教育厅科研基金资助项目(2006sk010)

摘　　要：蚁群优化是人工智能领域中群体智能分支之一,已成功地应用于复杂优化问题的求解,但其在知识发现领域的应用还是一个新的研究课题。在此提出一种新的基于蚁群优化的分类规则挖掘方法,先利用蚁群算法通过对属性约简简化数据集,再使用蚁群算法进行分类规则的挖掘,并用新的规则剪枝方法,提高了分类算法的效率和准确率。实验表明该方法是有效的。Ant Colony Optimization （ACO） is a branch of a newly developed form of artificial intelligence called swarm intelligence. It has shown its performance in solving complex optimization problems. But it is still a new research topic in Knowledge Discovery in Database （KDD）. This paper proposes a new method based on ACO of mining classification rule from categorical database. Firstly, it uses ACO to attribute reduction for minimizing dataset. Then it uses ACO to discover classification rule and redefine a new pruning procedure. Hence there are better efficiency and higher classification accuracy. The experiment shows that the proposed method is effective.

关 键 词：群体智能 蚁群算法 分类 属性约简 规则剪枝 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]