首頁 > 網路資源 > 大同大學數位論文系統

Title page for etd-0128108-172738


URN etd-0128108-172738 Statistics This thesis had been viewed 2299 times. Download 38 times.
Author Chih-Chiang Hsu
Author's Email Address No Public.
Department Information Management
Year 2007 Semester 1
Degree Master Type of Document Master's Thesis
Language zh-TW.Big5 Chinese Page Count 54
Title Fuzzy Particle Swarm Optimization for Data Clustering
Keyword
  • Fuzzy PSO
  • Particle swarm optimization
  • Fuzzy
  • Data Clustering
  • Data Clustering
  • Fuzzy
  • Particle swarm optimization
  • Fuzzy PSO
  • Abstract This paper proposes a new data clustering algorithm which is based on fuzzy techniques and particle swarm optimization (PSO). As pointed out by some researchers: the standard PSO always converges very quickly towards the optimal positions but may slow its convergence speed when it is reaching a minimum [9]. This paper is trying to solve this problem by integrating a Fuzzy technique with PSO to allow each particle to update its new velocity and next position according to the current position of other better particles, in addition to gbest, pbest and itself. Not all of other particles are considered by a particle; only those particles which have higher degree of fuzzy membership with the current global best particle are taken into account. This provides more information to direct the particle to fly toward a better direction. Finally, the proposed algorithm was evaluated by testing a couple of hard and soft clustering problems, also compared to some famous clustering methods. The experimental results show that the proposed approach has better convergence ability than its original PSO algorithm.
    Advisor Committee
  • Yen-Ju Yang - advisor
  • YuCheng Kao - advisor
  • Yi-Ching Chen - co-chair
  • Files indicate in-campus access only
    Date of Defense 2007-12-18 Date of Submission 2008-01-31


    Browse | Search All Available ETDs