Research on the Fuzzy Quantitative Association Rules Mining Algorithm and Its Simulation

Abstract

A key problem of mining quantitative association rules is to partition the continuous quantitative attribute. In this paper, it has been solved by using fuzzy partition, which can provide a smooth transition of data partition. Further more, based on the formal definition of fuzzy quantitative association rules, a quantitative association rules mining algorithm is proposed. This algorithm partitions continuous quantitative attribute using fuzzy clustering method to transform the original continuous quantitative attribute data into fuzzy membership function matrix, and then association rules can be mining. The simulation research based on large scale database shows that the mining algorithm of fuzzy quantitative association rules is effective and suitable for the quantitative association rules mining and knowledge discovery of large scale database.