AN EFFICIENT ALGORITHM FOR MINING FREQUENT ITEM-SETS CONTAINING A CONSTRAINT SUBSET

Abstract

For users, it is necessary to find frequent item-sets including a subset Co whose value is changed regularly. Recently, many studies showed that the mining of frequent item-sets with often changed constraints should be based on the lattice of closed item-sets and generators instead of mining them directly from the database. In this paper, a structure and unique representation of frequent item-sets restricted on constraint Co using closed frequent item-sets and their generators are proposed. Then, an algorithm called MFS-Contain-IC is developed to quickly and distinctly generate all frequent item-sets containing Co. The theoretical results are proven to be reliable. Extensive experiments on a broad range of synthetic and real data-sets show the effectiveness of present study.