Deriving General Association Rules from XML Data

XML documents have become poplar because the semi-structure nature of XML allows a wide variety of data to be represented in XML. Association rule mining is an important problem in the data mining domain. Currently, the problem of association rule mining on XML data has not been well studied. Existing work only addresses the problem of mining specific association rules from XML data. Such techniques specify antecedent and consequence to particular elements, and then mine rules with those specific antecedents and consequences. These techniques can not be used to mine general association rules. In this paper, we address the problem of deriving general association rules form XML data and propose an approach to perform the task. We implement our approach using Java DOM and test our algorithm on market basket data represented in XML