A fast algorithm for mining frequent patterns

Although mining based on FP-tree achieves high performance and efficiency, it still suffers from creating conditional FP-tree separately and recursively during the mining process. Moreover, update mining is very time-consuming. We propose a new method that designs a new structure called trans-tree, which stores all of the information in a highly compact form. The algorithm mines frequent patterns are based on trans-tree. Updating with new data needs one scan of the new data. Thus, it improves performance greatly.