首页 | 本学科首页   官方微博 | 高级检索  
     

基于局部密度比的模糊隶属度设置算法
引用本文:杨晓伟,,邵壮丰 梁艳春,,吴春国,. 基于局部密度比的模糊隶属度设置算法[J]. 吉林大学学报(理学版), 2006, 44(6): 41-44
作者姓名:杨晓伟    邵壮丰 梁艳春    吴春国  
作者单位:1. 华南理工大学 数学科学学院, 广州 510640; 2. 吉林大学 计算机科学与技术学院, 长春 130012;3. 吉林大学 符号计算与知识工程教育部重点实验室, 长春 130012
摘    要:基于知识发现中的局部异常因子思想, 提出一种基于局部密度比的模糊隶属度设置算法, 该算法根据样本的邻域密度比设置样本的隶属度, 并采用一种单参数选择策略. 数值实验表明, 所提出的算法在带噪声的非线性函数估计方面具有很好的鲁棒性, 有效地解决了模糊支持向量机中的模糊隶属度设置问题, 对处理带噪声的分类和非线性函数估计问题具有重要的意义.

关 键 词:模糊支持向量机  局部异常因子  局部密度比  模糊隶属度  
收稿时间:2006-06-02

A Local density ratio Based Algorithm for Setting Fuzzy Memberships
YANG Xiao wei,,SHAO Zhuang feng,LIANG Yan chun,,WU Chun guo,. A Local density ratio Based Algorithm for Setting Fuzzy Memberships[J]. Journal of Jilin University: Sci Ed, 2006, 44(6): 41-44
Authors:YANG Xiao wei    SHAO Zhuang feng  LIANG Yan chun    WU Chun guo  
Affiliation:1. School of Mathematical Sciences, South China University of Technology, Guangzhou 510640, China;2. College of Computer Science and Technology, Jilin University, Changchun 130012, China; 3. Key Laboratory ofSymbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun 130012, China
Abstract:Based on the local outlier factor (LOF) for detecting outlier in knowledge discovery, a local density ratio (LDR) based setting fuzzy membership algorithm was developed. In the proposed algorithm, the fuzzy memberships are assigned to the samples according to their neighborhood density ratios and a single parameter selection strategy is also adopted. Numerical experiments showed that the proposed algorithm possesses a good robustness for nonlinear function estimation problems with noise data. The presented algorithm is effective for setting fuzzy memberships in fuzzy support vector machine, which is important to deal with classification problems and nonlinear function estimation problems with noise data.
Keywords:fuzzy support vector machine  local outlier factor  local density ratio  fuzzy membership
点击此处可从《吉林大学学报(理学版)》浏览原始摘要信息
点击此处可从《吉林大学学报(理学版)》下载全文
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