Research of an association rule mining algorithm based on FP tree

Based on analyzing an association rule mining algorithm called FP tree, a new association rule mining algorithm called QFP was presented. Through scanning the database only once, the QFP algorithm can convert a transaction database into a QFP tree after data preprocessing, and then do the association rule mining of the tree. The QFP algorithm is more integrity than the FP_growth algorithm, and retain the complete information for mining frequent patterns; it will not destroy the long pattern of any transaction, and significantly reduce the non-relevant information. Experiments show that the QFP algorithm is more efficient if the aspect of time than the FP_growth algorithm.