A Hybrid Method for Frequent Itemsets Mining

Discovering frequent patterns is one of the essential topic data mining. A new algorithm based on the two-way-hybrid search for frequent itemsets mining is proposed. 1) A hierarchical search space organization is presented, based on which the original search space can be recursively decomposed into some smaller independent pieces. 2) A novel HFMI algorithm, which explores a flexible two-way-hybrid search method, is given. It executes the mining in both the top-down and bottom-up directions. Information gathered in the bottom-up can be used to prune the search space in the other top-down direction. Some efficient decomposition and pruning strategies are implied in this method, which can reduce the original search space rapidly in the iterations. 4) Experimental and analytical results are presented in the end of this paper.