A FP-Tree-Based Method for Inverse Frequent Set Mining

Abstract

Recently, the inverse frequent set mining problem has received more attention because of its important applications in different privacy preserving data sharing contexts. Several studies were reported to probe the NP-complete problem of inverse frequent set mining. However, it is still an open problem that whether there are reasonably efficient search strategies to find a compatible data set in practice. In this paper, we propose a FP-tree-based method for the inverse problem. Compared with previous “generation-and-test” methods, our method is a zero trace back algorithm, which saves huge computational costs. Furthermore, our algorithm provides a good heuristic search strategy to rapidly find a FP-tree, leading to rapidly finding the compatible databases. More importantly, our method can find a set of compatible databases instead of finding only one compatible database in previous methods.