多重Ki-ary Tree為架構之XML結構化索引

Abstract

近年來可延伸標記語言XML(Extensible Markup Language)已成為非常熱門的研究主題，許多文件格式都以XML為傳遞交換的基礎，XML可說是現今網際網路中儲存和傳送資訊最有發展之標準。而XML文件之中隱含了極多的結構化資訊，為了能使這些結構化資訊能提供結構化檢索(Structured Query)，本論文以K-ary 樹狀結構(K-ary Tree)為理論基礎，延伸發展出多重Ki-ary樹狀結構(Multiple Ki-ary Tree)，並以此理論建立XML文件之結構化索引(Structured Index)。本論文所提出的Multiple Ki-ary Tree的資料結構所佔的空間較K-ary Tree Split Method少。此外，由實驗得知：在不分裂的節點配置情況下，Multiple Ki-ary Tree在資料轉換的效能上比K-ary Tree為佳，且Multiple Ki-ary Tree能處理K-ary Tree所不能處理的某些特殊類型之XML文件。

In recent years XML(Extensible Markup Language) has become a very popular research topic and it has been used to be a basis of transmission and exchange for many document formats. For example, Taiwanese government has endeavored to promote electronic form of official documents and exchanges of electronic official documents in recent years, whose transmission standard are based on XML. XML could be said to be the most developed standard for storing and transmitting information on the Internet at present time. To create structured index and to provide structured query on XML document, this thesis thus utilizes characteristics of K-ary tree structure as theoretical basis to extend and develop Multiple Ki-ary Tree Structure and to establish a Structured Index of XML document through this theory.