Font Size: a A A

XML Query Based On Keyword

Posted on:2013-12-03Degree:MasterType:Thesis
Country:ChinaCandidate:J T GaoFull Text:PDF
GTID:2248330371972967Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
XML has become an important method of storing data, how to query XML data is currently a hot issue. There are two main ways of XML retrieval, one is using XML standard query language, such as XPath and XQuery. However, users need to know the grammar and the structure of underlying data. Thus, this manner is appropriate for professional users. The other is the XML Keyword Search, which only needs users to enter keywords of the query and will output the results to users. Because of its simplicity, it is applied to the majority of ordinary users.Researchers have deeply studied XML Keyword Search, but there do still exist the following problems:1. The first one is that keyword search is limited to a single XML document stored as a flat file, not to deal with the XML data stored in the relational database.2. The second one is the current approach haven’t cached keyword search results, which leads to wasting lots of time for repeat queries.3. The last one is the importance of evaluation of query result from users didin’t get much attention.To overcome the above drawbacks, we built a XML keyword search prototype system, called WXKS.This paper first introduces the knowledge of XML keyword search, including XML basic concepts, XML specifications and standards, XML tree structure, Dewey coding, LCA, XML path pattern and structured query language.Second, it describes WXKS in detail. In WXKS, the data for quering is the XML documents stored in the relation database. Quering results are including all the segments with root of LCA and present to user by webpages. Users browse the results, pick some one they really want and give rank for the results to the system. When a keyword query are firstly commtted to WXKS, the query results are cached in database. WXKS adjusts the order of the results in the database according to some rank method. Once a same keyword query are putted again by some users, WXKS gets the corresponding results from the database, thus, saves lots of time. We conduct experiments to verify the feasibility of WXKS, the experiment results show that WXKS can solve the problems mensioned above.
Keywords/Search Tags:XML keyword search, LCA, RANK, query result cache
PDF Full Text Request
Related items