Study of an improved Apriori algorithm for data mining of association rules

Abstract

Data mining of association rules provides the technology for discovering the interesting association or correlation from mass of data. Apriori algorithm can find all the frequent items from transactional databases, and eliminate non-frequent items. But, the Apriori algorithm for data mining of association rules always produces a large number of candidate items, and scans the database repeatedly. Z-Apriori algorithm, the improved Apriori algorithmfor data mining of association rules, is introduced. A numerical example about a supermarket is given to show that Z-Apriori algorithm can dig the weighted frequent items easily and quickly. The association rules and items which are more interested by customers and more profitable can be found by Z-Apriori algorithm, and they are also traditionally supported highly.