Algorithm Based on Bit Objects for Mining Maximal Frequent Patterns

A new algorithm based on bit objects,BFP-Miner,for mining maximal frequent patterns was proposed.It uses the bit objects to express data and to improve the FP-Tree(frequent pattern tree).The algorithm uses depth-first search strategy,and simplifies the support counting of frequent patterns with the characteristics of the bit objects and by compression of the database.Neither a conditional FP-Tree nor candidate patterns are generated during mining the maximal frequent patterns,so that the mining efficiency is increased.Experimental result verifies the efficiency of the BFP-Miner.