首頁 > 網路資源 > 大同大學數位論文系統

Title page for etd-0226115-171936


URN etd-0226115-171936 Statistics This thesis had been viewed 1119 times. Download 164 times.
Author Yu-Cheng Chen
Author's Email Address kevinchen1227@gmail.com
Department Information Management
Year 2014 Semester 2
Degree Master Type of Document Master's Thesis
Language zh-TW.Big5 Chinese Page Count 37
Title Crossover Strategic Particle Swarm Optimization with Local Search for Solving Data Clustering Problem
Keyword
  • Particle Swarm Optimization (PSO)
  • Evolution Computing (EC)
  • Swarm Intelligence (SI)
  • Hybrid Algorithm
  • Data Clustering
  • Data Clustering
  • Hybrid Algorithm
  • Swarm Intelligence (SI)
  • Evolution Computing (EC)
  • Particle Swarm Optimization (PSO)
  • Abstract Structured and unstructured data implicitly contain valuable vast information which utilize
    Decision Support System (DSS) and prediction data model for Business Intelligence (BI) in
    any fields of science and commercial business. Data classification and clustering problem both
    are very important research in data-mining technology. This paper aims at data clustering
    problem with a hybrid particle swarm algorithm intends to improve its performance. By which
    algorithm identifies similar properties or characters of data objects to aggregate different
    clusters. Our proposed CSPSO-LS algorithm mainly use PSO as its evolution framework with
    strategic crossover operation ideas from Genetic Algorithm, which hybrids Simulated
    Annealing Algorithm as local search method for solving data clustering optimization problem.
    In our experiments, the 5 datasets which we are intended to use are selected from the real cases
    of UCI are easily available at http://archive.ics.uci.edu/ml/datasets.html, in which are 5
    different dimensions and numbers of data objects. In experimental results of datasets for
    proposed algorithm, we compare with in nearly a decade of researchers, our proposed CSPSOLS
    algorithm comes out better performance as results. We also point some discussions and research direction on proposed CSPSO-LS algorithm.
    Advisor Committee
  • Huei-Huang Chen - advisor
  • Yu-Chen Kao - co-chair
  • Files indicate in-campus access immediately and off-campus access at one year
    Date of Defense 2015-01-14 Date of Submission 2015-03-02


    Browse | Search All Available ETDs