DTGC-Tree: A new strategy of association rules mining

Abstract

Efficient algorithms for mining frequent itemsets are crucial for mining association rules. As a condensed and complete representation of all the frequent itemsets, closed frequent itemsets mining has arisen a lot of interests in the data mining community. However, most of the studies havenpsilat addressed the effects of noise in the data sets on the algorithms, and there has been limited attention to the development of noise tolerant algorithms. In this paper, we represent a noise tolerant algorithm, DTGC-Tree, which based on an intuitive idea: applying association rules as soon as possible. By this way, the new algorithm could prune a lot of duplicated closed itemsets in the transactional data base. The performance evaluation demonstrates that the proposed algorithm could stand against noise and is both time and space efficient.