An Efficient Association Rule Mining for XML Data

XML association rule mining is an important problem in data mining domain. Currently, the problem of association rule mining on XML data has not been well studied. In this paper, we proposed an efficient association rule mining for large amount of XML data. The set of data is viewed as a binary table. The value of the itemset is one, if the corresponding XML data exist in the dataset, zero for otherwise. The frequent itemset is generated along with the candidate key. The closed itemset for the given data set is also generated. The closed itemset don't have any superset. For both frequent itemset and closed itemset generation we use Apriori algorithm. The possible association rules are generated for the XML data. Then the generated Association rule is converted into XML format. Our proposed system EARM may reduce the memory storage size and it returns association rules with short response time.