The arithmetic of discover Cyclic Association Rules based on difference sequence arithmetic clustering

Abstract

The routine Cyclic Association Rules and its upgrading alternatives were found with some problems as these:Some disadvantage of technique that compartmentalize a cycle into several time segments; The basic arithmetic use Apriori Arithmetic ,Their disadvantage are huge the candidate items and low-level efficiency. In regard to these problems, a new Cyclic Association Rules method was discussed. The new approach addressed these problems by adopting different methods.It chose the time sequence vector consists of the support of item to cluster, and using DB Index to determine the Optimal Class Number of Cluster, accordingly confirm well and truly time segments of cycle. And we introduced Cyclic FP-tree(CFP-tree) to discover Cyclic Association Rules, Cyclic FP-tree based on FP-tree arithmetic,FP-tree arithmetic excelled evidently Apriori arithmetic, CFP-tree handle cycle clipping technology based on conditional FP-tree to improve markedly efficiency of arithmetic. The findings tested Data Set of high-dimensional industrial authenticity. The findings prove the efficiency of the new approach in a broad scale.