淡江大學機構典藏:Item 987654321/54540
English  |  正體中文  |  简体中文  |  全文笔数/总笔数 : 62830/95882 (66%)
造访人次 : 4076570      在线人数 : 803
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU 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


    jsp.display-item.identifier=請使用永久網址來引用或連結此文件: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/54540


    题名: 多階層遺傳模糊探勘技術之研究
    其它题名: Multi-Level Genetic-Fuzzy Data-Mining Techniques
    作者: 陳俊豪
    贡献者: 淡江大學資訊工程學系
    关键词: 資料探勘;遺傳演算法;多階層模糊關聯規則;隸屬函數;遺傳模糊探勘;Data mining;Genetic algorithm;Multi-level fuzzy association rules;Membership functions;Genetic-fuzzy mining
    日期: 2010
    上传时间: 2011-07-06 11:51:36 (UTC+8)
    摘要: 遺傳模糊探勘的議題主要是結合遺傳演算法與模糊理論從交易資料中同時探勘隸屬函數與模糊關聯規則的技術。故架構上分為兩個步驟，分別為隸屬函數探勘與模糊關聯規則探勘步驟。在第一步驟首先使用演化式計算找出適合於探勘問題的隸屬函數，之後再使用最後的最佳隸屬函數去探勘模糊關聯規則。可惜的是，截至目前為止，上述遺傳模糊資料探勘研究成果僅針對一般的數值型交易資料，對於具概念階層的數值型交易資料庫，仍未見有研究進行探討。 本計畫的主要目的在發展多階層遺傳模糊探勘方法，同時探勘商品最小支持度、隸屬函數與多階層模糊關聯規則。根據遺傳模糊探勘技術應注重的因素，包括：隸屬函數型態、規則有效性、最小支持度合適性與語意項目個數合適性，本計畫提出一個兩年的研究，針對下列的研究議題進行探討： (1) 在單一最小支持度模糊關聯規則探勘(SMSFM)問題下，針對不同的多階層模糊關聯規則模式，包括：逐層探勘模式(Level-by-level mode)與跨層探勘模式(Generalized mode)，設計發展單一最小支持度之多階層遺傳模糊探勘技術。 (2) 在多重最小支持度模糊關聯規則探勘(MMSFM)問題下，考量不同的最小支持度限制(Minimum Support Constraints)策略與多階層模糊關聯規則模式，設計發展多重最小支持度之多階層遺傳模糊探勘技術。
    Genetic-fuzzy data mining techniques are used to discover membership functions and useful fuzzy association rules by combining the genetic algorithms and the fuzzy concepts. The framework consists of two phases, including mining membership function and mining fuzzy association phases. In the first phase, the genetic algorithm is used to derive appropriate membership functions for items. In second phase, the final membership functions are then used to mine fuzzy association rules. Unfortunately, all of the literature on genetic-fuzzy data mining, to our best knowledge, is confined to the quantitative transaction database environment; no research work has been conducted on genetic-fuzzy data mining over quantitative transaction database with taxonomy. The aim of this project is to develop the multi-level genetic-fuzzy data mining methods for mining minimum supports, membership functions and multi-level fuzzy association rules from quantitative transaction database with taxonomy. According to the general issues in designing genetic-fuzzy data mining algorithms, such as types of membership functions, the effectiveness of the rules, suitability of minimum supports and appropriate number of linguistic terms, in this context, we will propose a two-year project, focusing on the following main issues: (1) For single-minimum-support fuzzy-mining problem, we attempt to design multi-level genetic-fuzzy data mining techniques for items with a single minimum support based on two types of multi-level rule mining modes, including level-by-level and generalized modes. (2) For multiple-minimum-support fuzzy-mining problem, we attempt to design multi-level genetic-fuzzy data mining techniques for items with multiple minimum supports based on two types of multi-level rule mining modes, and by considering two types of minimum support constraints.
    显示于类别:[資訊工程學系暨研究所] 研究報告

    文件中的档案:

    没有与此文件相关的档案.

    在機構典藏中所有的数据项都受到原著作权保护.

    TAIR相关文章

    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - 回馈