XML資料庫搜尋技術之研發

Abstract

此論文之研究重點為XML資料庫之搜尋與擷取技術，希望能掌握此一最新技術，增加往後對XML-based資料之執行效益。對目前XML技術中與XQuery相關的如Schema與Xpath作一介紹，同時將目前對XML文件具有搜尋機制的XML DOM、XSLT與ＸＱuery作一比較。 現今，為了儲存和管理XML的文件，有兩種主要的處理方式，一是儲存在關聯式資料庫中，再者就是儲存在新發展的原生型XML資料庫(Native XML database)中。至於何種方式較為適合現今之商務資訊系統，截至目前為止，還沒有定論。雖然目前已有許多研究人員積極投入相關的研究，迄今在原生型XML資料庫中的效能評估仍然不完整。 本網站系統以二層式資料結構的題庫系統為實例，以XML文件為資料庫，ASP & DOM的相互合作，為發展網頁的工具，研究XML文件資料庫的應用。由研究發現XML文件直接當資料庫使用可減少資料轉換關聯式資料庫的負擔，但為顧及查詢與擷取資料的效率，資料庫的儲存需考慮儲存檔案的適當大小與文件結構不可太複雜，以提高網路的效率及提高查詢的速度，經過實驗驗証了此系統之功能及有效性。

This thesis focused on the search and retrieval of XML database system. An introduction about XML technology related to XQuery such as Schema and XPath. is presented. At the same time, a comparison is made between XML DOM and XSLT, which are major search engines for XML documents. Nowadays, there are two ways to store and manage XML documents. One way is to store in the “relation database”; the other is to store in the newly-developed “Native XML database”. Up to now, there is no conclusion which one is a better way to apply to contemporary commerce information system. Although many researchers have gone into this kind of study, the functional evaluation of the Native XML database is still incomplete. A prototype XML database system is developed using examination question system as an example. A two-tier client server architecture is used.. Interactive cooperation of ASP and DOM are used to develop web pages and to study the application of XML database. From this study, we can find that using XML documents can reduce the task of translating data into relational database. However, considering the efficiency in searching and retrieval the size of the XML documents should be properly designed and the structure of XML documents should not be too complicated. Through experiments, we validate the functionalities and effectiveness of the system.