In the traditional closed frequent itemsets mining on transactional databases, which items have no weight (equal weight, as equal to 1). However, in real world applications, usually, each item has a different weight (the importance/significance of each item). Therefore, we need to minine weighted/significance itemsets on transactional databases. In this paper, we propose a high-performance algorithm called HPM-CFSI for mining closed frequent significance itemsets based on approach NOT satisfy the downward closure property (a great challenge). The experimental results show that the proposed algorithms perform better than other existing algorithms on both reallife and synthetic datasets.
Closed frequent significance itemsets, High-performance, HPMCFSI algorithm, Downward closure property