Research On Key Technology Of Entity-based XML Keyword Search Processing

XML, stands for eXtensible Markup Language, is a self-describing and extensible language. XML has become the de facto standard for data representation and exchange for Web applications. With the spread of application of XML, more and more data are stored and exchanged in the form of XML. XML Keyword Query can help users to retrieve information they need from amount of XML documents, only needs users input several keywords,so the technology of XML Keyword Query has been a hot research area of XML. In this paper, we aim at solving the problems of existing XML Keyword Query techniques, the main research of this paper are as follows:Firstly, in order to resolve the problems of not returning any meaningful results and returning meaningless results, our paper classifies the XML nodes and introduces the entity based semantics, namely ELCEA, which returns only entity nodes, entity nodes are meaningful as results. Our paper also improves the ELCEA to capture users'search intension by supporting additionally the semantics of"NOT". In our paper, stack algorithms are proposed to process the ELCEA semantic efficiently.Secondly, our paper analyzes the queries which contain NOT predicate, defines the notion of the minimum result tree, also designs construction algorithm of minimum result tree, then analyzes the present research condition of XML Keyword rank, also considers the structure and semantic information of XML to discuss the rank strategy of XML Keyword Query, .Rich experiments are conducted on both real and synthetic datasets. Our paper introduces the data sources and experimental development environment. Then demonstrates the query effectiveness towards precision and recall metric, also demonstrates the query efficiency of our methods through the query time.