International Journal of Computer Science and Mobile Computing (IJCSMC)
In this paper, the authors describe an approach to mining tree-based association rules from XML documents. Such rules provide information on both the structure and the content of XML documents; moreover, they can be stored in XML format to be queried later on. The mined knowledge is approximate, intentional knowledge used to provide: quick, approximate answers to queries and information about structural regularities that can be used as data guides for document querying. A prototype of the proposed system is also briefly described.