政大機構典藏-National Chengchi University Institutional Repository(NCCUR):Item 140.119/61318
English  |  正體中文  |  简体中文  |  Post-Print筆數 : 27 |  全文笔数/总笔数 : 109951/140892 (78%)
造访人次 : 46201150      在线人数 : 537
RC Version 6.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
搜寻范围 查询小技巧：
  • 您可在西文检索词汇前后加上"双引号"，以获取较精准的检索结果
  • 若欲以作者姓名搜寻，建议至进阶搜寻限定作者字段，可获得较完整数据
    •  进阶搜寻


    请使用永久网址来引用或连结此文件: https://nccur.lib.nccu.edu.tw/handle/140.119/61318


    题名: 多分類器系統建構之理論與實務
    作者: 徐國偉
    Hsu,Kuo-Wei
    贡献者: 國立政治大學資訊科學系
    关键词: 多分類器系統;分類器差異性;分類器整合
    Multiple Classifier System;Multi-Classifier System;Ensemble;Diversity;Classifier Combination
    日期: 2012
    上传时间: 2013-10-11 15:13:07 (UTC+8)
    摘要: 本計畫旨在發展多分類器系統建構之理論與實務。近年來在
    資料探勘領域，多分類器系統是一個越來越有吸引力的研究
    主題且被應用在各式各樣的領域。然而，我們對其基礎仍然
    缺乏全面了解。我們可能知道哪個多分類器系統是有效的，
    但是我們卻不清楚它為什麼有效，而且我們對於如何有效地
    建構一個有效的多分類器系統的知識是有限的。雖然有許多
    成功的案例，但是在更多案例裡面多分類器系統不被認為是
    成功的。因此，能夠解釋為何一個多分類器系統是不成功的
    與能夠解釋為何一個多分類器系統是成功的是同等重要。再
    者，大多數多分類器系統採取的是針對特定問題或資料的特
    別設計，然而這樣的設計嚴格限制了他們可能的延伸與應
    用。如何設計一個泛用型的多分類器系統是一個重要的研究
    課題。在本計畫中，我們將採用理論與實證研究去對多分類
    器系統的建構做深入的探討，並聚焦於以異質分類演算法建
    置的多分類器系統。本計畫可對國內外的多分類器系統研究
    與應用做出實質貢獻，而部分研究成果已發表於專業國際會
    議。
    The purpose of this project is to develop theory and
    practice of the construction of Multiple Classifier
    System. Recently, in the data mining field, Multiple
    Classifier System has been an increasingly attractive
    research topic and been applied to various domains.
    However, we still lack a comprehensive understanding
    of its foundation. We may know which Multiple
    Classifier System is effective but we are not clear
    why it is effective, and we have limited knowledge of
    how to effectively construct an effective Multiple
    System. Although there are many success cases, there
    are even more cases where Multiple Classifier System
    is not considered success. Therefore, being able to
    explain why a Multiple Classifier System is
    unsuccessful is equally important to being able to
    explain why a Multiple Classifier System is
    successful. Furthermore, most systems are ad-hoc
    designs, i.e. they are designed specifically to some
    problems or data sets, while such designs severely
    restrict their possible extensions and applications.
    For that reason, a critical research topic is how to
    design a general-purpose Multiple Classifier System.
    In this project, we plan to use theoretical and
    empirical study to have an in-depth of the
    construction of Multiple Classifier System, and we
    will focus on the Multiple Classifier Systems built
    on heterogeneous classification algorithms. This
    project could make practical contributions to
    research and applications of Multiple Classifier
    System, and part of the results have been published
    on professional international conferences.
    關聯: 執行期間：10008-10107
    数据类型: report
    显示于类别:[資訊科學系] 國科會研究計畫

    文件中的档案:

    档案 描述 大小格式浏览次数
    100-2218-E-004-002.pdf2325KbAdobe PDF21160检视/开启


    在政大典藏中所有的数据项都受到原著作权保护.


    社群 sharing

    著作權政策宣告 Copyright Announcement
    1.本網站之數位內容為國立政治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，惟仍請適度，合理使用本網站之內容，以尊重著作權人之權益。商業上之利用，則請先取得著作權人之授權。
    The digital content of this website is part of National Chengchi University Institutional Repository. It provides free access to academic research and public education for non-commercial use. Please utilize it in a proper and reasonable manner and respect the rights of copyright owners. For commercial use, please obtain authorization from the copyright owner in advance.

    2.本網站之製作，已盡力防止侵害著作權人之權益，如仍發現本網站之數位內容有侵害著作權人權益情事者，請權利人通知本網站維護人員(nccur@nccu.edu.tw)，維護人員將立即採取移除該數位著作等補救措施。
    NCCU Institutional Repository is made to protect the interests of copyright owners. If you believe that any material on the website infringes copyright, please contact our staff(nccur@nccu.edu.tw). We will remove the work from the repository and investigate your claim.
    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - 回馈