A Novel Fuzzy Positive and Negative Association Rules Algorithm

According to the existing mining algorithm of fuzzy association rules, a novel fuzzy positive and negative association rules algorithm will be proposed in this paper. We focus on the membership function of fuzzy set and minimum support parameters of positive and negative association rules and adopt a method that selects parameters automatically which is based on the k-means clustering. Besides, multi-level fuzzy support and correlation coefficient are chosen to restrain the quantity and quality of rules generated by the algorithm. Finally the validity and accuracy of the algorithm are proved by an experiment.