Key Technology For Multidimensional Negative Association Rules Mining

In recent years, the research on the multidimensional association rules is mainly for the positive association rules. The research on multidimensional negative association rules is relatively small. In this case, the algorithm of multidimensional negative association rules mining is important.Although laying a more mature (negative) association rule mining algorithm theoretical basis.But the multidimensional negative association rules mining in this area have not yet found in domestic and foreign research results and related literature. We propose an algorithm designed for the multi-dimensional positive and negative association rules. We build a multi-dimensional test positive and negative association rule mining algorithms for specific areas of the system, and to test the feasibility of the algorithm on computer class registration database University of correlation analysis.Multidimensional association rules aim to find positive and negative rules, but some rules which have been found are meaningless. So, in the framework of support– confidence, the third parameter which aims to delete redundant rules is introduced. In such a situation, the multidimensional negative association rules mining algorithms have been proposed.This research work on multi-dimensional association rules further provide a theoretical basis for the research and application basis.We propose an algorithm for multidimensional positive and negative association rules mining. We proposed B-ai algorithm to implement with the code. For example, we achieve simultaneously positive and negative rule mining. We need multidimensional data to import into the program, this multidimensional datas are digitized, we lead into the minimum support and minimum confidence, then using algorithms to multidimensional data set is divided into two major items infrequent itemsets and frequent itemsets.In the frequent itemset,we have mining multidimensional positive association rules, In the infrequent itemset, we have mining multidimensional negative association rules. In the infrequent itemset, we have multidimensional mining negative association rules.We completed the design of algorithms and the corresponding theoretical analysis and experimental simulation of the performance of the algorithm are discussed, show that the algorithm is feasible and effective.