计算机工程 ›› 2011, Vol. 37 ›› Issue (19): 44-46.doi: 10.3969/j.issn.1000-3428.2011.19.013

• 软件技术与数据库 • 上一篇    下一篇

基于RPCL的模糊关联规则挖掘

谢 皝,张平伟,罗 晟   

  1. (上海大学计算机工程与科学学院,上海 200072)
  • 收稿日期:2011-03-21 出版日期:2011-10-05 发布日期:2011-10-05
  • 作者简介:谢 皝(1986-),男,硕士,主研方向:数据挖掘,文本挖掘;张平伟,教授;罗 晟,硕士

Fuzzy Association Rule Mining Based on Rival Penalized Competitive Learning

XIE Huang, ZHANG Ping-wei, LUO Sheng   

  1. (School of Computer Engineering and Science, Shanghai University, Shanghai 200072, China)
  • Received:2011-03-21 Online:2011-10-05 Published:2011-10-05

摘要: 在模糊关联规则的挖掘过程中,很难预先知道每个属性合适的模糊集。针对该问题,提出基于次胜者受罚竞争学习的模糊关联规则挖掘算法,无需先验知识,即可根据每个属性的性质找出对应的模糊集,并确定模糊集的数目。实验结果表明,与同类算法相比,该算法可以挖掘出更多有趣的关联规则。

关键词: 模糊集, 隶属度函数, 聚类, 次胜者受罚竞争学习, 关联规则

Abstract: It is very hard to know in advance the most appropriate fuzzy sets that are good enough for the domains of quantitative attributes for fuzzy association rules mining. So this paper proposes a fuzzy association rule mining algorithm based on Rival Penalized Competitive Learning (RPCL), which can get fuzzy sets and set the number of fuzzy sets without prior knowledge of the fuzzy linguistic terms but by quantitative attributes. Experimental results show that the algorithm can get more interesting association rules compared with other algorithms.

Key words: fuzzy set, membership function, clustering, Rival Penalized Competitive Learning(RPCL), association rule

中图分类号: