Robust solutions to fuzzy one-class support vector machine

被引:9
作者
Liu, Yong [1 ,2 ]
Zhang, Biling [1 ,2 ]
Chen, Bin [1 ]
Yang, Yandong [1 ]
机构
[1] Beijing Univ Posts & Telecommun, Sch Network Educ, Beijing 100088, Peoples R China
[2] Xidian Univ, State Key Lab Integrated Serv Networks, Xian 710071, Peoples R China
基金
中国国家自然科学基金;
关键词
One-class SVM; Fuzzy system; Robustness;
D O I
10.1016/j.patrec.2015.12.014
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
One-class SVM is used for classification which distinguishes one class of data from the rest in the feature space. For the training samples coming from different sources with different quality, in this letter, a reformulation of one-class SVM is proposed by simultaneously incorporating robustness and fuzziness to improve the classification performance. Based on the proposed model, we derive the relationship between the lower bound of fuzziness mu(min) and the upper bound of perturbation eta in the input data. Specifically, for a given eta, only when the assigned fuzziness to the input data is larger than mu(min) could the input data be in full use and differentiated effectively. The experiments verify the mathematical analysis and illustrate that the proposed model can achieve better classification performance. (C) 2016 Elsevier B.V. All rights reserved.
引用
收藏
页码:73 / 77
页数:5
相关论文
共 12 条
  • [1] Chang C.-C., 2004, Libsvm: a library for support vector machines, software
  • [2] Applying the Possibilistic c-Means Algorithm in Kernel-Induced Spaces
    Filippone, Maurizio
    Masulli, Francesco
    Rovetta, Stefano
    [J]. IEEE TRANSACTIONS ON FUZZY SYSTEMS, 2010, 18 (03) : 572 - 584
  • [3] Fuzzy one-class support vector machines
    Hao, Pei-Yi
    [J]. FUZZY SETS AND SYSTEMS, 2008, 159 (18) : 2317 - 2336
  • [4] Huang G., 2012, IEEE T NEURAL NETW L, V23
  • [5] Kjærgaard MB, 2007, LECT NOTES COMPUT SC, V4718, P139
  • [6] Fuzzy support vector machines
    Lin, CF
    Wang, SD
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 2002, 13 (02): : 464 - 471
  • [7] An introduction to kernel-based learning algorithms
    Müller, KR
    Mika, S
    Rätsch, G
    Tsuda, K
    Schölkopf, B
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 2001, 12 (02): : 181 - 201
  • [8] Primal-dual interior-point methods for self-scaled cones
    Nesterov, YE
    Todd, MJ
    [J]. SIAM JOURNAL ON OPTIMIZATION, 1998, 8 (02) : 324 - 364
  • [9] Estimating the support of a high-dimensional distribution
    Schölkopf, B
    Platt, JC
    Shawe-Taylor, J
    Smola, AJ
    Williamson, RC
    [J]. NEURAL COMPUTATION, 2001, 13 (07) : 1443 - 1471
  • [10] Support vector data description
    Tax, DMJ
    Duin, RPW
    [J]. MACHINE LEARNING, 2004, 54 (01) : 45 - 66