基于版本树的改进XML枝匹配算法研究-计算机应用技术专业论文.docxVIP

AbstractABSTRACT Abstract ABSTRACT At present，XML(Extensible Markup Language)has become the standard for data representation and data exchange over World Wide Web．As more and more XML data emerges，query processing over XML data shows its increasing importance．Currently,a large number of XML query processing algorithms based on tree model have been successively proposed．However,semi-structured nature of XML document and increasing complex of the users’queries bring new challenges OVer efficient processing XML data as follows：(1)the input size and execution time of these algorithms will mcrease rapidly as the size of XML document increases；(2) due to the irregular nature of XML data，the results that algorithm retumed may exist duplicate phenomenon and require post-processing operation．Therefore，how to efficiently and accurately execute query on a large number of XML data becomes an important topic． In this paper,we propose an efficient and extensible XML twig pattern matching algorithm named Twi93Version and an efficient XML twig pattern matching algorithm named AdvancedTwigVersion in case of the returned results are matched twigs and non-redundant target elements respectively．The main research work and characteristic of the thesis are aS follows： (1)In case of the returned results are matched twigs，we propose an idea of applying the proposed holistic twig pattern matching algorithms based on original document to compressed structure(Version Tree)，present simple and efficient version filter module and me玛e join module．Structure matching on the concise Version Tree along with version filter and merge join on the concise intermediate results greatly reduce the query time． (2)In case of the returned results are non-redundant target elements，we propose AdvancedTwigVersion algorithm which improves TwigVersion algorithm． By using the label information of the ancestor element as much as possible，those elements that do not satisfy match can be filtered out as early aS possibl