Construction of periodic temporal association rules in data mining

Abstract

Association rules are the very valuable kind of law in data mining. The fitness of time is seldom illustrated by traditional association rules, which losses a number of useful implicit rules. On the basis of further study of other association rules mining algorithms, this paper has developed Apriori-extended mining periodic temporal association rules (MPTAR) according to the especial periodicity of data. The test on a group of financial data shows that the method is useful and efficient. It is more significant for improving the theory and implementation of temporal association rules in data mining.