PRIVACY PRESERVING FREQUENT ITEMSET MINING

Abstract

In today’s world, large amount of data is used in organizations around the world. This large amount of data needs to be processed so that we can extract useful information. In data mining, most important algorithm for finding frequent itemset from a large amount of data is Apriori Algorithm. The Apriori algorithm wastes a lot of time for scanning the whole database and finding the frequent itemsets. Additionally, the existing techniques in most of the research papers for hiding sensitive pattern is static. But nowadays, most of the data that is received by an organization is dynamic in nature. This research work points toward the inadequacy of the Apriori algorithm of wasting time for scanning the whole transactional database for finding frequent patterns. Hash Link Based approach is proposed which works on progressive data in order to hide sensitive itemsets from a set of frequent itemsets in progressive data. This improved algorithm uses hash links in order to reduce the scanning time for whole database and hence improved the time complexity by great extent.