首页 | 本学科首页   官方微博 | 高级检索  
     

一种新的基于粗集编码的模糊聚类数据处理方法
引用本文:曾黄麟,师奕兵,曾晓辉.一种新的基于粗集编码的模糊聚类数据处理方法[J].电子测量与仪器学报,2007,21(3):39-42.
作者姓名:曾黄麟  师奕兵  曾晓辉
作者单位:四川理工学院,自贡,643000;电子科技大学,成都,610054;成都信息工程学院,成都,610225
基金项目:四川省教育厅基础应用研究课题
摘    要:本文提出一种新的基于粗集编码的模糊聚类数据处理方法.该方法对电子测量信息处理中的数据,根据粗集理论进行编码、特征属性简化,然后利用模糊隶属度函数将输入精确信息映射为模糊变量信息,提出把数据特征的重要性因子结合在模糊聚类的分类隶属度函数中以提高数据聚类处理的能力,并利用最小化目标函数离线学习来搜索测量数据聚类的聚类中心,该方法可以通过人工神经网络实现.

关 键 词:模糊  粗集  神经网络  聚类
修稿时间:2006-04

New Method of Data Processing Based on Fuzzy Clustering of Rough Set Encoding
Zeng Huanglin,Shi Yibing,Zeng Xiaohui.New Method of Data Processing Based on Fuzzy Clustering of Rough Set Encoding[J].Journal of Electronic Measurement and Instrument,2007,21(3):39-42.
Authors:Zeng Huanglin  Shi Yibing  Zeng Xiaohui
Affiliation:Sichuan University of Science and Engineering, 643033, P. R. China; University of Electronic Science and Technology of China, 610054, P.R. China; Chengdu University of Information Technology, 610225, P. R. China
Abstract:A new approach of data processing based on fuzzy clustering of rough set encoding is presented m this paper. An equivalence class encoding input data is defined to eliminate insignificant feature attributes in data sets of electronic measurement data processing by means of rough sets. Fuzzy representation of precise input data is used to deal with either incomplete or imprecise even ill -defined database. A class membership function incorporated the significant factor of the input feature attribute is made to enhance data processing characteristic corresponding to consequent class in the fuzzy clustering output space. One kind of supervised algorithms with batch expression is suggested in searching for the cluster center of data classification by way of LMS rule. The method proposed here can be realized by way of an artificial neural network.
Keywords:fuzzy set  rough set  neural network  clustering  
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号