Pfp

• DOI: 10.1145/1454008.1454027
• Publication Date: 2008-11-06T13:49:50Z
• Authors: Haoyuan Li; Yi Wang; Dong Zhang; Ming Zhang; Edward Y. Chang
• Keywords: Speedup; Empirical Research

Frequent itemset mining (FIM) is a useful tool for discovering frequently co-occurrent items. Since its inception, number of significant FIM algorithms have been developed to speed up performance. Unfortunately, when the dataset size huge, both memory use and computational cost can still be prohibitively expensive. In this work, we propose parallelize FP-Growth algorithm (we call our parallel PFP) on distributed machines. PFP partitions computation in such way that each machine executes an independent group tasks. Such partitioning eliminates dependencies between machines, thereby communication them. Through empirical study large 802,939 Web pages 1,021,107 tags, demonstrate achieve virtually linear speedup. Besides scalability, demonstrates promising supporting query recommendation search engines.