National Taiwan Ocean University Institutional Repository:Item 987654321/28643
English  |  正體中文  |  简体中文  |  Items with full text/Total items : 28611/40652
Visitors : 775647      Online Users : 50
RC Version 4.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Adv. Search

Please use this identifier to cite or link to this item:

Title: Validation-Driven Algorithm for SVC
Authors: Keng-Hsuan Wu;Jung-Hua Wang
Contributors: NTOU:Department of Electrical Engineering
Date: 2005
Issue Date: 2011-10-21T02:38:45Z
Publisher: IEE Proceedings - Vision, Image and Signal Processing
Abstract: abstract:This paper presents a novel approach, calledvalidation-driven algorithm (VDA) for improving supportvector clustering (SVC). The SVC algorithm is a well-knownkernel-based clustering approach, but its clustering resultskeenly rely on the proper choice of parameters of kernelfunctions. Particularly, we have found that performance ofSVC is very sensitive to the initial value of the kernelparameter qini, and a poor initial q value inevitably will incurheavy computation time for obtaining a satisfactory result.Furthermore, it is very difficult to determine the proper qvalue for input containing clusters with large diversity invariations. Previous SVC algorithms use an identical trial-and-error q for all clusters. Aiming to tackle these issues, VDA notonly capable of finding a good initial value of the q parameter,but also provides a computationally efficient procedure forverifying the validity of the parameter setting as well as theclustering result. Simulation results demonstrate theeffectiveness of the proposed VDA algorithm.
Relation: pp.782-787
Appears in Collections:[Department of Electrical Engineering] Periodical Articles

Files in This Item:

File Description SizeFormat

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


著作權政策宣告: 本網站之內容為國立臺灣海洋大學所收錄之機構典藏,無償提供學術研究與公眾教育等公益性使用,請合理使用本網站之內容,以尊重著作權人之權益。
網站維護: 海大圖資處 圖書系統組
DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback