本研究之目的為定義一個新的Web探勘系統框架（Web Mining System Framework, WMSF），設計出經濟的Web探勘架構，開發出一個用戶化全球資訊網探勘系統（Customized W3 Mining System, CWMS），以降低構建Web探勘平台的成本。然後驗證與評估CWMS的Web爬蟲效率與彈性的可調節性與可移植性。本研究方法是將文獻探討所得之理論模式做一整合，並作成研究架構。其理論之推論以Web探勘系統框架的定義、用戶化全球資訊網探勘系統、CWMS的設計與實現、CWMS實驗研究為主。在本研究中，依據Cha等人研究的基於關鍵字的文本數據收集和分析的集成框架的功能體系結構中之Web爬蟲組件與使用者介面組件中的結果視圖面板（Result View Panel, RVP），提出了一個新的Web探勘系統框架（WMSF），並依照Oh等人研究的Script腳本執行和地址提取的過程（虛擬演算法），有效地應用PHP語言開發了用戶化全球資訊網探勘系統（CWMS），該系統組件包括演算法程式、巨量儲存空間和Web伺服器。最後，在CWMS的實驗研究結果已驗證了該系統可以提高Web探勘效率與彈性的可調節性與可移植性。;The aim of this study is to define a new web mining system framework (WMSF), to design an economical web mining architecture, to develop a customized W3 mining system (CWMS) for the cost down on the web mining platform. Then, we verify and evaluate the web crawler efficiency of CWMS and the adjustability and portability of resiliency. The method of the research is to integrate the theoretical model obtained from the literature and to make a research structure. The inference of the theory is with the definition of WMSF, the CWMS, the design and implementation of CWMS and the experimental research of CWMS as a focus. In this study, according to the functional architecture of integrated framework for keyword-based text data collection and analysis studied by Cha et al., and we extract the web crawler component and the result view panel (RVP) of user interface component in the architecture and then reorganize a new WMSF. And then, according to the script execution and address extraction process (virtual algorithm) studied by Oh et al., the effective using PHP language to develop the CWMS which is components include algorithm programs, huge amounts storage space and web server. Finally, in the experimental results of CWMS, we have verified that the system can improve the web crawler efficiency of CWMS and the adjustability and portability of resiliency.
