Mining associations based on simplified concept lattice by improved algorithm
基于约简概念格的关联规则提取改进算法*

Concept lattice is an important technique of mining association rules. Generating all the large itemsets for the concept lattice need sort all the nodes of the lattice and then compare them one by one. A new algorithm based on simplified concept lattice is proposed to improve the efficiency of generating large set form the concept lattice. The algorithm can generate all the large itemsets by using the parents-children correlation of the nodes. It improves the efficiency of generating large sets by cutting the time cost of sorting nodes and reducing the comparing times between the nodes. Through the experiments on performance study, it can be seen that the proposed algorithm has much superior performance in efficiency to the original algorithm.