结合模糊(C+P)均值聚类和SP-V-支持向量机的TSK分类器

徐明亮; 王士同

doi:10.3788/OPE.20162403.0643

您当前的位置：

首页 >

文章列表页 >

结合模糊(C+P)均值聚类和SP-V-支持向量机的TSK分类器

信息科学 | 更新时间：2020-08-13

- 结合模糊(C+P)均值聚类和SP-V-支持向量机的TSK分类器
- TSK classifier based on fuzzy(C+P) means clustering and SP-V-SVM
- 光学精密工程 2016年24卷第3期页码：643-650
- 作者机构：
  
  1. 无锡环境科学与工程研究中心,江苏无锡,214063
  2. 江南大学数字媒体学院, 江苏无锡 214122
- 作者简介：
- 基金信息：
  
  国家自然科学基金资助项目(No.61202311,No.61272210)();江苏省青蓝工程资助项目(2014)()
- DOI：10.3788/OPE.20162403.0643
  中图分类号： TP391
- 收稿日期：2015-12-08，
  
  修回日期：2016-01-14，
  
  纸质出版日期：2016-03-25
- 稿件说明：
移动端阅览
徐明亮, 王士同,. 结合模糊(C+P)均值聚类和SP-V-支持向量机的TSK分类器[J]. 光学精密工程, 2016,24(3): 643-650

XU Ming-Liang, WANG Shi-Tong,. TSK classifier based on fuzzy(C+P) means clustering and SP-V-SVM[J]. Editorial Office of Optics and Precision Engineering, 2016,24(3): 643-650
徐明亮, 王士同,. 结合模糊(C+P)均值聚类和SP-V-支持向量机的TSK分类器[J]. 光学精密工程, 2016,24(3): 643-650 DOI： 10.3788/OPE.20162403.0643.

XU Ming-Liang, WANG Shi-Tong,. TSK classifier based on fuzzy(C+P) means clustering and SP-V-SVM[J]. Editorial Office of Optics and Precision Engineering, 2016,24(3): 643-650 DOI： 10.3788/OPE.20162403.0643.

摘要

为获得具有模糊规则自适应约简性能和较好的泛化性能的TSK分类器

本文提出了一种结合模糊(C+P)均值聚类(FCPM)算法和SP-V-支持向量机(SVM)分类算法来构建TSK(Takagi-Sugeno-Kang)分类器的方法。该方法首先用FCPM聚类算法对训练数据进行聚类;然后根据聚类结果确定TSK分类器的模糊规则前件中的高斯隶属度函数的中心和宽度参数;最后采用成组稀疏约束SP-V-SVM算法对模糊规则后件参数进行学习

该算法不仅改善了系统的泛化性能

还使系统具有模糊规则自适应约简功能

使得系统更为紧凑。与相关算法在UCI和IDA标准数据集分类实验中的模糊规则数和分类性能对比表明:用提出的分类算法所构造的TSK分类器不仅具有较好的分类性能

而且模糊规则数少

有利于构建更为紧凑的模糊分类系统。

Abstract

To obtain a TSK(Takagi-Sugeno-Kang) classifier with well generalization performance and a capability of rule adaptive deduction

the method based on the FCPM(Fuzzy(C+P) Means) clustering and the SP-V-SVM(Support Vector Machine) is proposed to construct the TSK classifier. The method firstly clusters train data by FCPM clustering algorithm

and then identifies the centers and widths of the Gaussian membership functions of the antecedent part in TSK classifier based on the clustering result. Finally it uses the SP-V-SVM algorithm with the group sparsity constraints to learn the parameters of the consequent part of the fuzzy rule to improve the generalization performance. Moreover

the attribute reduction capacity of the SP-V-SVM algorithm is used to reduce the number of fuzzy rules and to make the TSK classifier compacted. The classification error and the number of fuzzy rules of the proposed algorithm are compared with several standard data sets from UCI and IDA repository

and results show that the proposed TSK classifier not only gets better performance but also has a few number of fuzzy rules and is suitable for structuring compact fuzzy classifying systems.

关键词

Keywords

references

常湛源, 曹军, 张怡卓. 板材心理感知颜色在线模糊分类器设计[J]. 电机与控制学报, 2014, 18(9):93-98.CHANG ZH Y, CAO J, ZHANG Y ZH. Design of on-line timber plate fuzzy classification fused with psychological perceptions of color[J], Electri c Machines and Control, 2014, 18(9):93-98.(in Chinese)

GUILLAUME S. Designing fuzzy inference systems from data:an interpretability oriented review[J]. IEEE Transactions on Fuzzy Systems, 2001, 9(3):426-443.

DAHAL K, ALMEJALLI K, HOSSAIN M A, et al.. GA-based learning for rule identification in fuzzy neural networks[J]. Applied Soft Computing, 2015, 35:605-617.

MAHDIZADEH M, EFTEKHARI M. Generating fuzzy rule base classifier for highly imbalanced datasets using a hybrid of evolutionary algorithms and subtractive clustering[J]. Journal of Intelligent and Fuzzy Systems, 2014, 27(6):3033-3046.

邢宗义, 张永, 侯远龙, 等. 基于模糊聚类和遗传算法的具备解释性和精确性的模糊分类系统设计[J]. 电子学报, 2006, 34(1):83-88.XING Z Y, ZHANG Y, HOU Y L et al.. Design of interpretable and precise fuzzy classification system based on fuzzy clustering and genetic algorithm[J]. ACTA ELECTRONICA SINICA, 2006, 34(1):83-88.(in Chinese)

王莉, 周献中, 李华雄. 基于决策粗糙集的模糊分类模型[J]. 信息与控制, 2014, 43(1):24-29.WANG L, ZHOU X ZH, LI H X. Fuzzy classification model based on decision-theoretic rough set[J]. Information and Control, 2014, 43(1):24-29.(in Chinese)

JUANG C F, CHIU S H, and CHANG S W. A self-organizing ts-type fuzzy network with support vector learning and its application to classification problems[J]. IEEE TRANSACTIONS ON FUZZY SYSTEMS, 2007, 15(5):998-1008.

郭敬明, 何昕, 魏仲慧. 基于在线支持向量机的Mean Shift彩色图像跟踪[J]. 液晶与显示, 2014, 29(1):120-128.GUO J M, HE X, WEI ZH H. New mean shift tracking for color image based on online support vector machine[J]. Chinese Journal of Liquid Crystals and Displays, 2014,29(1):120-128.(in Chinese)

汪方斌, 洪津, 孙晓兵, 等. 基于支持向量机预测偏振模式[J]. 光学精密工程, 2014, 22(11):2914-2922.WANG F B, HONG J, SUN X B, et al.. Prediction of polarization pattern by SVM[J]. Opt. Precision Eng., 2014, 22(11):2914-2922.(in Chinese)

王丽, 魏巍, 吴林钢, 等. SAR图像目标识别新方法[J]. 液晶与显示, 2014, 29(3):429-434.WANG L, WEI W, WU L G et al.. Novel target recognition method for SAR[J]. Chinese Journal of Liquid Crystals and Displays, 2014, 29(3):429-434.(in Chinese)

LIN C T, YEH C M, LIANG S F, et al.. Support-vector-based fuzzy neural network for pattern classification[J]. IEEE TRANSACTIONS ON FUZZY SYSTEMS, 2006, 14(1):31-41.

CHEN Y X, WANG J Z. Support vector learning for fuzzy rule-based classification systems[J]. IEEE TRANSACTIONS ON FUZZY SYSTEMS, 2003, 11(6):716-728.

CHIANG J H, HAO P Y. Support vector learning mechanism for fuzzy rule-based modeling:a new approach[J]. IEEE TRANSACTIONS ON FUZZY SYSTEMS, 2004, 12(1):1-12.

CHAVES A D F, VELLASCO MMBR, TANSCHEITR. Fuzzy rules extraction from support vector machines for multi-class classification[J]. Neural Computing and Applications, 2013, 22(7-8):1571-1580.

蔡前凤, 郝志峰, 刘伟. 基于模糊划分和支持向量机的TSK模糊系统[J]. 模式识别与人工智能, 2009, 22(3):411-416.CAI Q F, HAO ZH F, LIU W. TSK fuzzy systems based on fuzzy partition and support vector machines[J]. PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE, 2009, 22(3):411-416.(in Chinese)

LESKI J M, Fuzzy(c + p)-means clustering and its application to a fuzzy rule-based classifier:towards good generalization and good interpret-ability[J]. IEEE TRANSACTIONS ON FUZZY SYSTEMS, 2015, 23(4):802-812.

VANESSA G V, MANEL M R, JERÓNIMO A G et al.. Support vector machines with constraints for sparsity in the primal parameters[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 2011, 22(8):1269-1283

浏览量

3552

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据