A perfect hashing to enhance the performance of Apriori algorithm

Abstract

Data mining is the process of discovering patterns, relationships, and insights from large datasets using various computational techniques. Association Rules is a data mining method to find various relations between items in an itemset. The Apriori algorithm is a popular and classical algorithm for association rule mining. It has been observed that an increase in the number of candidate sets decreases the efficiency of the Apriori algorithm rapidly. To overcome this issue, the hashing technique, which uses a hash function to reduce the size of the candidate set itemset, is often used. In this paper, a perfect hashing function for a d-itemset is proposed. The efficiency of the Apriori algorithm is enhanced by using the hash function.