English  |  正體中文  |  简体中文  |  Items with full text/Total items : 25397/25992 (98%)
Visitors : 6189185      Online Users : 36
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version


    Please use this identifier to cite or link to this item: http://ir.lib.ksu.edu.tw/handle/987654321/19621


    Title: 建構在核心函數之修正山峰分類模型-運用聚類整合技術並探討其應用
    Authors: 吳國龍
    鄭乃嘉
    林宥均
    Keywords: 核函數
    改良山峰法
    聚類整合
    強韌性
    相關性資料
    類別資料
    圖像分割
    人臉檢測
    聚類
    聚類有效性
    Kernel function
    Modified mountain method
    Cluster ensembles
    Robust
    Relational data
    Categorical data
    Image segmentation
    Face detection
    Clustering
    Cluster validity
    Date: 2012-11-02
    Issue Date: 2013-07-15 17:01:28 (UTC+8)
    Publisher: 行政院國科會:專題研究計畫
    Abstract: 在這一研究專案的第一年,我們將結合'核函數'和'改良山峰法'的概念來創建基於核函數的改良山峰分類模型。我們將討論一些特別的核函數和其相對應的影子核函數。核函數的頻寬選擇問題將會用聚類整合技巧來探討並估計。新的分類模式的性能將會與其他聚類方法如K-均值,模糊k-均值和EM演算法進行比較。我們將探討不同核函數的影響函數與損壞點特性並與並與其他聚類方法相比。我們預期新的模式可以解決初始化、強韌性和群集有效性的問題。
    In the first year of this research project, we will combine the concepts of 'kernel functions' and 'modified mountain methods' to create a kernel based modified mountain clustering model. Some special kernels and their shadows will be discussed. The bandwidth selection problems for the kernels will be analyzed and estimated by the cluster ensemble technique. The performance of the proposed model will compare with other clustering methods, such as K-means, fuzzy k-means and EM algorithm. We will also establish the influence function and breakdown point of the proposed method. The robust properties of different kernels will be obtained and compared to other clustering methods. We expect the proposed model can solve the problems of initializations, robustness and cluster validity.
    Appears in Collections:[資訊管理系所] 研究計畫

    Files in This Item:

    File Description SizeFormat
    report-12.pdf1688KbAdobe PDF375View/Open


    All items in KSUIR are protected by copyright, with all rights reserved.


    本網站之所有圖文內容授權為崑山科技大學圖書資訊館所有,請勿任意轉載或擷取使用。
    ©Kun Shan University Library and Information Center
    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback