Abstract
With the growth of a large amount of marketing data, mining useful and meaningful association rules from databases has become an important research topic. Previous works have focused on the attributes of the marketing data to derive association rules. In this paper, we consider the attributes of transactions and allow users to specify queries against the attributes and then discover the interesting correlations among the derived association rules. For efficiency, we organize the marketing data as a multiple-attribute hierarchical tree by the attributes of transactions to derive the corresponding association rules. Finally, we make experiments on a synthetic database for performance evaluation and comparisons.