An efficient distributed algorithm for mining association rules

Association Rule Mining (ARM) is an active data mining research area. However, most ARM algorithms cater to a centralized environment where no external communication is required. Distributed Association Rule Mining (DARM) algorithms aim to generate rules from different datasets spread over various geographical sites; hence, they require external communications throughout the entire processor. A direct application of sequential algorithms to distributed databases is not effective, because it requires a large amount of communication overhead. DARM algorithms must reduce communication costs. In this paper, a new solution is proposed to reduce the size of message exchanges. Our solution also reduces the size of average transactions and datasets that leads to reduction of scan time, which is very effective in increasing the performance of the proposed algorithm. Our performance study shows that this solution has a better performance over the direct application of a typical sequential algorithm.