• DocumentCode
    3339256
  • Title

    A combination of mixture Genetic Algorithm and Fuzzy C-means Clustering Algorithm

  • Author

    Liu, Su-Hua ; Hou, Hui-Fang

  • Author_Institution
    Coll. of Comput. Sci., Wuhan Univ. of Technol., Wuhan, China
  • Volume
    1
  • fYear
    2009
  • fDate
    14-16 Aug. 2009
  • Firstpage
    254
  • Lastpage
    258
  • Abstract
    Firstly, the paper makes a briefly analysis and comment about the fuzzy c-means clustering algorithm. Then a new kind of hybrid genetic algorithm is proposed on the base of the combination of genetic algorithm and simulated annealing algorithm, and it is applied in fuzzy c-means clustering. It overcomes the locality and the Sensitivity to initial clustering central of fuzzy c-means clustering, by using randomness and parallelism in hybrid genetic algorithm searching. And a new tree-shaped coding scheme adapted to fuzzy clustering is adopted in the genetic algorithm. In the end, the paper supplies the detailed design of the method. Simulation experiments show the relatively high efficiency and recognition accuracy of the method, which has extensive application prospect in many fields, such as pattern recognition, data mining, and so on.
  • Keywords
    fuzzy set theory; genetic algorithms; pattern clustering; simulated annealing; trees (mathematics); fuzzy c-means clustering algorithm; hybrid genetic algorithm; simulated annealing algorithm; tree-shaped coding scheme; Clustering algorithms; Clustering methods; Computer science; Genetic algorithms; Genetic engineering; Information analysis; Information science; Paper technology; Pattern recognition; Simulated annealing; Fuzzy Clustering; Genetic Algorithm; Simulated Annealing Algorithm; Tree-shaped coding;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    IT in Medicine & Education, 2009. ITIME '09. IEEE International Symposium on
  • Conference_Location
    Jinan
  • Print_ISBN
    978-1-4244-3928-7
  • Electronic_ISBN
    978-1-4244-3930-0
  • Type

    conf

  • DOI
    10.1109/ITIME.2009.5236422
  • Filename
    5236422