Mining intensional information for answering XML-queries using tree-based association rules approach

Nowadays web pages are mostly XML based and most of this information does not have well-defined Document Type Definition (DTD) or Schema. They are relatively huge document and incomplete in nature. Therefore effective answering of the XML document is tedious as it leads to irrelevant and more generic results thereby failed to offer the specific user´s needs. To avoid this problem, an approach based on Tree-based Association Rule (TAR) mining was proposed, using which for efficient answering the frequent paths and references of its occurrence will be extracted and it is stored in XML format for both ordered and unordered XML document. The constructed XML file is used for query answering; thereby it provides quick and approximate answers to user´s query. The user´s query is automatically transferred into a query format that retrieves information from the constructed XML file.