Volume 8 Number 5 (May 2013)
Home > Archive > 2013 > Volume 8 Number 5 (May 2013) >
JSW 2013 Vol.8(5): 1101-1108 ISSN: 1796-217X
doi: 10.4304/jsw.8.5.1101-1108

A Feature Weighted Spectral Clustering Algorithm Based on Knowledge Entropy

Hongjie Jia1, Shifei Ding1, 2, Hong Zhu1, Fulin Wu1, Lina Bao1

1School of Computer Science and Technology, China University of Mining and Technology, Xuzhou, China
2Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, 100190 China


AbstractSpectral clustering has aroused extensive attention in recent years. It performs well for the data with arbitrary shape and can converge to global optimum. But traditional spectral clustering algorithms set the importance of all attributes to 1 as default, when measuring the similarity of data points. In fact, each attribute contains different information and their contributions to the clustering are also different. In order to make full use of the information contained in each attribute and weaken the interference of noise data or redundant attributes, this paper proposes a feature weighted spectral clustering algorithm based on knowledge entropy (FWKE-SC). This algorithm uses the concept of knowledge entropy in rough set to evaluate the importance of each attribute, which can be used as the attribute weights, and then applies spectral clustering method to cluster the data points. Experiments show that FWKE-SC algorithm deals with high-dimensional data very well and has better robustness and generalization ability.

Index TermsSpectral clustering, rough set, knowledge entropy, attribute importance.

[PDF]

Cite: Hongjie Jia, Shifei Ding, Hong Zhu, Fulin Wu, Lina Bao, "A Feature Weighted Spectral Clustering Algorithm Based on Knowledge Entropy," Journal of Software vol. 8, no. 5, pp. 1101-1108, 2013.

General Information

  • ISSN: 1796-217X (Online)
  • Frequency:  Quarterly
  • Editor-in-Chief: Prof. Antanas Verikas
  • Executive Editor: Ms. Yoyo Y. Zhou
  • Abstracting/ Indexing: DBLP, EBSCO, CNKIGoogle Scholar, ProQuest, INSPEC(IET), ULRICH's Periodicals Directory, WorldCat, etc
  • E-mail: jsweditorialoffice@gmail.com
  • APC: 500USD
  • Jun 12, 2024 News!

    Vol 19, No 2 has been published with online version   [Click]

  • Jan 04, 2024 News!

    JSW will adopt Article-by-Article Work Flow

  • Apr 01, 2024 News!

    Vol 14, No 4- Vol 14, No 12 has been indexed by IET-(Inspec)     [Click]

  • Apr 01, 2024 News!

    Papers published in JSW Vol 18, No 1- Vol 18, No 6 have been indexed by DBLP   [Click]

  • Mar 01, 2024 News!

    Vol 19, No 1 has been published with online version    [Click]