Generating 3rd Level Association Rules Using Fast Apriori Implementation

Abstract

The mining of association rules and frequent item sets are the main area of interest in recent research activities. The multiple level association rules provide the more meaningful information in comparison to single level association rules which describes only single level concept hierarchy. The Apriori algorithm is most established algorithm for mining the single level association rules. In this study, the fast implementation of Apriori algorithm has been used to generate 3rd level association rules with some modifications in the algorithm. The data coding and data cleaning techniques are used to find the multilevel association rules as they are prerequisite to implement the modified algorithm.