Title: Improved Frequent Pattern Mining Algorithm Based on FP-Tree
Abstract: The FP-Growth algorithm is currently one of the most efficient frequent pattern mining algorithms, which does not produce a large number of candidate item sets in the process of data mining.However, the traditional FP-Growth algorithm needs to scan twice conditional pattern bases to construct the FP-Tree and conditional FP-Tree, which runs slowly and consumes a lot of time.To optimize this algorithm, an improved FP-Growth algorithm is put forward, which is carried out by using a two-dimensional table to record support counts between items, therefore the second traversal of conditional pattern bases can be avoided.In this paper, some instances and experiments are implemented.Experimental results demonstrate that this improved FP-Growth algorithm based on a two-dimensional table is an efficient frequent pattern mining algorithm of higher performance than the traditional one.