Feature Selection by Principle Component Analysis for Mining Frequent Association Rules

Data mining techniques have been increasingly studied. Extracting the association rules have been the focus of this studies. Recently research have focused on association rules to help uncover relationships between seemingly unrelated data in a relational database or other information repository. The large size of data makes the extraction of association rules hard task. In this paper, we propose a new method for dimension reduction and feature selection based on the Principal Component Analysis, then find the association rules by using the FP-Growth Algorithm. Experimental results reveals that the reduction technique can discover the same rules obtained by the original data.