An Efficient Frequent Pattern Mining using Boolean Matrix and Transaction Reduction Technique

Abstract: Association Rule Mining is one of the most important functionality in data mining, by using that we can find out frequent patterns. Here we propose a novel method of finding frequent patterns by using Boolean matrix for transactional data set and finding the support count by using “Logical AND ” operation and here we can also reduce the transactions by the assumption “ A Transaction which do not contain frequent k-items can not contain frequent k+1-items in it. So such transactions can be removed from the transactional data set. So that we can save the memory space and reduce the computational time to generate frequent patterns