Parallel Mining Association Rules in Calculation Grids

As a result of the physical storage expansion and backup equipment and the increasing need to store more data, the sequential searching algorithms of association rules have been found ineffective. Thus the introduction of new parallel versions has become a necessity. We propose in this paper, a parallel version of a sequential algorithm Partition. This is fundamentally different from other sequential algorithms, as it scans the database only twice to generate all significant association rules. Consequently, the parallel approach does not require much communication between sites. The proposed approach was implemented for an experimental study. The results obtained show a large gain in execution time compared to the sequential version.