A New Algorithm for Parallel Association Rule Mining in Distributed Shared Memory System

Finding the frequent itemset is the most important problem in Association Rule Mining (ARM) because it is the most time costly step in ARM. In the case of huge number of transaction and items in the database, it is important to investigate efficient distributed algorithm for mining association rules. The efficient distributed algorithm must be scalable, easy partitioned and distributed of centralized database, minimize the calculation and communication. In this paper we present a new algorithm for finding frequent itemset, that is called HVPFI. This algorithm is an extension of our previous work. We then analyze the algorithm and compare it with other published algorithms.