Dimensionality reduction for association rule mining with IST-EFP algorithm

Frequent itemset generation is the important phase on association rule mining. With frequent itemset dataset, association rules will be obtained. The main problems that exist in association rule mining is the use of large computer main memory at the time of the formation of Frequent Itemset. EFP algorithms (Expand FP-Growth) overcome this problem by utilizing secondary storage as a processing area to store it in the table object in the database. Data management processes in a database done by using a DBMS (Database Management System). The database used is Oracle database which has its integrated DBMS. Table object is a representation of the set on set theory in mathematics. One of set theory type is an intersection, the result of intersection of a set will be smaller than originally set (dimensionality reduction). IST-EFP algorithm apply the concept of intersection of set theory in EFP algorithm that can reduce 2.33% more items while maintaining association rules obtained.