Mining Generalized Knowledge from Transaction Databases

Abstract

Mining association rules is an important task for knowledge discovery. We can analyze past transaction data to discover customer behaviors such that the quality of business decision can be improved. Various types of association rules may exist in a large database of customer transaction. The strategy of mining g association rules focuses on discovering large itemsets which are groups of items which appear together in a sufficient number of transactions. In this paper, we propose a graph-based approach to discover generalized association rules from a large database of customer transactions. This approach is to construct an association graph to indicate the associations between items, and then traverse the graph to generate large itemsets. Empirical evaluations show that our algorithm outperforms other algorithms which need to make multiple passes over the database.