An improved Fp-tree algorithm

Abstract

Fp-tree algorithm plays an important role in mining maximal frequent patterns and searching association rules. However, because of recursive function call and more pointer fields of each node, traditional Fp-tree algorithm brings about large memory usage and low efficiency. This paper proposed an improved Fp-tree algorithm which used a non-recursive mechanism to reduce the use of the stack memory, and reduced a pointer field of each node to save memory usage. Experimental results showed that the improved algorithm is more efficient in the case of large volumes of data. © 2011 Springer-Verlag Berlin Heidelberg.