Discovery of Quantitative Association Rules from a Large Database of Sales Transactions

Abstract

In this paper, we examine the issue of mining quantitative association rules in a large database of sales transactions. When purchased quantities are considered, Most of the supports for items associated with their purchased quantities may be low, and the number of potentially interesting association rules discovered may be few. In order to discover more potentially interesting rules, we present two partition algorithms to partition all the possible quantities into intervals for each item. We also propose an efficient mechanism to discover all the large itemsets form the partition result. Experimental results show that by our approach, the total execution time can be reduced significantly. Moreover, the number of potentially interesting association rules discovered from our partition result is larger than that of rules discovered from the original data, which demonstrates the significance of our work.