Association rules parallel algorithm based on FP-tree

FP-tree algorithm is the best algorithm in association rules algorithm, it dose not need to scan the database more times, and it dose not need to generate the candidate item set. Therefore, it has repeatedly been quoted by many researchers, and make more depth study based on it and they have made a lot of good results. But when faced with large-scale database, the effect is still not very good, In this paper, we propose a parallel algorithm based on parallel mechanism, we divide the large-scale data into many blocks and the blocks are allocated to different computer nodes for processing. By the way, we found it is very good in time efficiency, and it is fit for any large database to data mining.