计算机科学

计算机科学 ›› 2019, Vol. 46 ›› Issue (8): 249-254.doi: 10.11896/j.issn.1002-137X.2019.08.041

• 人工智能 • 上一篇    下一篇

基于阴影集的截集式可能性C-均值聚类截集门限的选取

雒僖, 范九伦, 于海燕, 梁丹   

  1. (西安邮电大学通信与信息工程学院 西安710121)
    (电子信息勘验应用技术公安部重点实验室 西安710121)
    (陕西省无线通信与信息处理技术国际合作研究中心 西安710121)
  • 收稿日期:2018-07-09 出版日期:2019-08-15 发布日期:2019-08-15
  • 通讯作者: 范九伦(1964-),男,教授,博士生导师,主要研究方向为模式识别与图像处理、模糊信息处理、图像安全技术研究,E-mail:jiulunf@xupt.edu.cn
  • 作者简介:雒僖(1995-),女,硕士生,主要研究方向为模式识别、聚类分析,E-mail:1450085678@qq.com;于海燕(1982-),女,博士,副教授,主要研究方向为模式识别与图像处理;梁丹(1993-),女,硕士生,主要研究方向为图像处理
  • 基金资助:
    国家自然科学基金项目(61671377,61571361,61601362),西安邮电大学西邮新兴团队(xyt2016-01)

Selection of Cutset Threshold for Cutset-type Possibilistic C-means Clustering Based on Shadowed Set

LUO Xi, FAN Jiu-lun, YU Hai-yan, LIANG Dan   

  1. (School of Communication and Information Engineering,Xi’an University of Posts and Telecommunications,Xi’an 710121,China)
    (Key Laboratory Electronic Information Application Technology for Scene Investigation,Ministry of Public Security,Xi’an 710121,China)
    (Internation Joint Resesrch Center for Wireless Communication and Information Processing,Xi’an 710121,China)
  • Received:2018-07-09 Online:2019-08-15 Published:2019-08-15

PDF (PC)

795

摘要： 截集式可能性C-均值聚类算法通过引入截集门限,修改典型性值,克服了可能性C-均值聚类算法的最关键问题:一致性聚类。针对算法中截集门限的选取问题,采用阴影集理论,提出了一种新的截集门限的选取方法。该算法利用最优化方法为每一个类确定一个阴影集阈值,并将该阈值作为截集门限;通过分析该选取方法对典型性值和中心偏移量的影响来改进典型性值的修改方式。最后,通过人工数据分析了新的截集门限选取方式对聚类算法性能的影响,利用实际UCI数据分析算法的迭代次数和聚类正确率。实验结果表明,给出的截集门限选取方法能够有效减少迭代次数,提高聚类正确率。

关键词: 截集门限, 截集式可能性C-均值聚类, 聚类核, 可能性C-均值聚类, 阴影集

Abstract: By introducing the cutset threshold and modifying the typicality,the cutset-type possibilistic C-means clustering algorithm overcomes the most critical problem (consistent clustering)of the possibilistic C-means clustering algorithm.Aiming at the parameter selection problem in the algorithm,this paper proposed a new method based on the shadowed set.This algorithm uses the optimization method to determine the threshold of the shadowed set for each cluster and takes this threshold as the cutset threshold.The modification method of the typicality is improved by analyzing the influence of the selection method on the typicality and the center deviation.Finally,the influence of the new parameter selection method on the performance of the clustering algorithm is analyzed by artificial dataset.The number of iterations and the clustering accuracy of the algorithm are analyzed through the UCI dataset.Experimental results show that the proposed method can effectively reduce the number of iterations and improve the accuracy of clustering

Key words: Cluster core, Cutset threshold, Cutset-type possibilistic C-means clustering, Possibilistic C-means clustering, Shadowed sets

中图分类号: 

  • TP311.13
[1]BEZDEK J C.Pattern Recognition with fuzzy objective function algorithms[M].New York:Plenum Press,1981.
[2]KRISHNAPURAM R,KELLER J M.A possibilistic approach to clustering[J].IEEE Transactions on Fuzzy Systems,1993,1(2):98-110.
[3]KRISHNAPURAM R,KELLER J M.The possibilistic C-means algorithm:insights and recommendations[J].IEEE Transactions on Fuzzy Systems,1996,4(3):385-393.
[4]TIMM H,BORGELT C,DÖRING C,et al.An extension to possibilistic fuzzy cluster analysis[J].Fuzzy Sets and Systems,2004,147(1):3-16.
[5]FERRARO M B,GIORDANI P.On possibilistic clustering with repulsion constraints for imprecise data[J].InformationScie-nces,2013,245:63-75.
[6]PAL N R,PAL K,KELLER J M,et al.A Possibilistic Fuzzy c-Means Clustering Algorithm[J].IEEE Transactions on Fuzzy Systems,2005,13(4):517-530.
[7]ASKARI S,MONTAZERIN N,FAZEL Z M H,et al.Genera- lized entropy based possibilistic fuzzy C-Means for clustering noisy data and its convergence proof[J].Neurocomputing,2017,219:186-202.
[8]SARKAR J P,SAHA I,MAULIK U.Rough Possibilistic Type-2 Fuzzy C-Means clustering for MR brain image segmentation[J].Applied Soft Computing,2016,46:527-536.
[9]XIE Z P,WANG S T,CHUNG F L.An enhanced possibilistic c-means clustering algorithm EPCM[J].Soft Computing,2008,12(6):593-611.
[10]HAMASUNA Y,ENDO Y.Sequential Extraction by Using Two Types of Crisp Possibilistic Clustering[C]∥Proceedings ofIEEE International Conference on Systems,Man,and Cybernetics.New York:IEEE Press,2013:3505-3510.
[11]XENAKI S D,KOUTROUMBAS K D,RONTOGIANNIS A A.Sparsity-Aware Possibilistic Clustering Algorithms[J].IEEE Transactions on Fuzzy Systems,2016,24(4):1611-1626.
[12]KOUTROUMBAS K D,XENAKI S D,RONTOGIANNIS A A.On the Convergence of the Sparse Possibilistic C-Means Algorithm[J].IEEE Transactions on Fuzzy Systems,2018,26(1):324-337.
[13]YU H Y,FAN J L.Cutset-type possibilistic c-means clustering algorithm[J].Applied Soft Computing,2018,64:401-422.
[14]PEDRYCZ W.Shadowed sets:representing and processing fuzzy sets[J].IEEE Transactions on Systems,Man,and Cybernetics,Part B (Cybernetics),1998,28(1):103-109.
[15]PEDRYCZ W.From fuzzy sets to shadowed sets:interpretation and computing[J].International Journal of Intelligent Systems,2009,24(1):48-61.
[16]ZHOU J,PEDRYCZ W,MIAO D Q.Shadowed sets in the cha-racterization of rough-fuzzy clustering[J].Pattern Recognition,2011,14(8):1738-1749.
[17]ZHANG K,KONG W R,LIU P P,et al.Partition region-based suppressed fuzzy C-means algorithm[J].Journal of Systems Engineering and Electronics,2017,28(5):996-1008.
[18]WANG H L,SHE K,ZHOU M T.Shadowed Sets-based Rough Fuzzy Possibilistic C-means Clustering[J].Computer Science,2013,40(1):191-194.(in Chinese) 汪海良,佘堃,周明天.基于阴影集的粗糙模糊可能性C均值聚类算法[J].计算机科学,2013,40(1):191-194.
[1] 汪海良，佘 堃，周明天.
基于阴影集的粗糙模糊可能性C均值聚类算法
Shadowed Sets-based Rough Fuzzy Possibilistic C-means Clustering
计算机科学, 2013, 40(1): 191-194.
[2] 郭晋华,苗夺谦,周杰.
基于阴影集的粗糙聚类闭值选择
Shadowed Sets Based Threshold Selection in Rough Clustering
计算机科学, 2011, 38(10): 209-210.
[3] .
基于改进的PCM支持向量描述多类分类器

计算机科学, 2008, 35(8): 149-153.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发