A Frequent Pattern Tree Algorithm for Mining Association Rule Using Genetic Algorithm

Abstract

In recent years, data mining is an important aspect for generating association rules among the large number of itemsets. Association rule mining is one of the techniques in data mining that has two subprocess. First, the process called as finding frequent itemsets and second process is association rules mining. In this subprocess, the rules with the use of frequent itemsets have been extracted. Researchers developed a lot of algorithms for finding frequent itemsets and association rules. The frequent pattern technique only used for very large dataset and it takes large memory space tree creation. The major advantage of using Genetic Algorithm is that it perform global search and the time complexity is less compared to other algorithms. In this paper, first, GA is used to optimize the large dataset. Second, the improved frequent pattern tree is used to mine the frequent itemset without generating conditional FP-tree.