ASIA unversity:Item 310904400/9662
English  |  正體中文  |  简体中文  |  全文笔数/总笔数 : 94286/110023 (86%)
造访人次 : 21659191      在线人数 : 502
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


    jsp.display-item.identifier=請使用永久網址來引用或連結此文件: http://asiair.asia.edu.tw/ir/handle/310904400/9662


    题名: 柔性計算法應用於資料庫行銷流程之績效評估—創新性產品之個案研究
    作者: 陳永信
    贡献者: 管理學院
    經營管理學系
    关键词: 資料庫行銷;柔性計算法;資料探勘;需求區隔
    Database marketing;soft computing;data-mining;needs-based segmentation
    日期: 2008
    上传时间: 2010-05-14 07:25:24 (UTC+0)
    摘要: 了解顧客並滿足其需求是行銷管理成功因素之一,有系統地收集個別顧客或有望顧客的詳細資訊並建立顧客資料庫就能推動資料庫行銷,展開接觸以瞭解其需求,有利於顧客關係的維持,並為他們開發出客製化產品以促成產品/服務的銷售。顧客資料庫的大量既存資料或從顧客資料衍生出來的市場調查初級資料,經由資料探勘的運作就可擷取出顧客需求區隔以及行為預測等有用資訊。除了傳統的多變量分析統計方法之外,以人類固有的認知、理解、學習、記憶、判斷機制為基礎而演繹出來的各種柔性計算方法,例如類神經網路、模糊邏輯及基因演算法也可用於資料探勘。

    柔性計算的商業應用研究文獻,大多是以個別方法對應某個案要求,甚少聯合使用多種方法來解決一貫性的整體議題。因此本研究順應美國加州柏克萊大學 Zadeh 教授的呼籲,整合類神經網路、模糊邏輯和基因演算法,嘗試建立資料庫行銷流程中涵蓋顧客需求區隔以至購買行為預測為止的方法論,並以創新性資訊通信產品的開發個案研究佐證柔性計算法的績效評估,期望對行銷管理學術領域提出創新的方法論,做出實質貢獻。

    本研究預期研究成果如下:
    (1) 報告本研究聯合使用多種柔性計算方法來解決資料庫行銷中一貫性整體議題的成果及績效評估。
    (2) 以業界實際資料驗證本研究之實務可行性。
    (3) 發表國際研討會論文。
    (4) 發表論文於知名國際期刊。
    Understanding customers in the market and satisfying their needs are among crucial success factors in marketing management. The capability for a company to collect comprehensive information of individual incumbent customers and prospects, and store it into a database empowers the company’s potency in maintaining relationship with customers and selling products/services via database marketing activities. It does so by integrating customer database and product database to elaborate building relationship, contacting, customizing offerings, and dealing with customers and finally leads to a transaction. Massive information in customer database as well as primary data rendered by a market survey from the existing customer base enables the execution of data-mining task to extract knowledge about customers’ needs segmentation and purchase behavior trend.

    Apart from conventional multivariate statistical approach, soft computing techniques like artificial neural network, fuzzy logic and genetic algorithm that are derived from human biological instinct such as cognition, learning, memory and judgment are also applicable for data-mining task. However, most soft computing research literatures in business application are manifested in the form of single-technique for single-case approach, rather than congruently combining several techniques to offer the solution that can tackle the contextual-apt problems that always happen in the real world business environment. This arouses the motivation in responding the urge of Professor Zadeh (UC Berkeley) to study how to present an innovative methodology that are supported by the consortium of various soft computing techniques, and justify its feasibility in the area of database marketing activities all the way from customer needs identification, segmentation and future purchase behavior prediction. A case study on a development project for information/communication product of radical innovation will be presented to evaluate the contribution of this to-be research work.

    The expected outcomes of this research work are listed as follows:
    (1) A presentation of the research outcome thanks to the consortium of various soft computing techniques as well as the performance evaluation report.
    (2) Justification for the feasibility of this research work via an industry survey data set that could be acquired through an information/communication product development project.
    (3) Presentation of the research work finding in the international academic conference.
    (4) Contribution of research paper to the highly-acclaimed international academic journal.
    显示于类别:[經營管理學系 ] 科技部研究計畫

    文件中的档案:

    档案 描述 大小格式浏览次数
    97陳永信1.doc22KbMicrosoft Word460检视/开启
    97陳永信2.doc25KbMicrosoft Word325检视/开启
    index.html0KbHTML425检视/开启


    在ASIAIR中所有的数据项都受到原著作权保护.


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