The Research And Application Of Multi-dimensional Association Rules Algorithm Based On Mondrian

Association rule mining is to find associated and useful knowledge in a large number of data sets , it is not only the first research of data mining problems,but also an important research direction in data mining.Multidimensional association rule is an important branch of accociation rule mining.Multidimensional association rule continue to be widely used in the business world. Thus, Multidimensional association rule has been rapid development and has become a research hotspot in recent years.Meanwhile, in recent years, people research multi-dimensional association rules algorithm in use of OLAP technology, and find out large number of multi-dimensional association rule algorithms based on OLAP data cubes . The use of multi-dimensional OLAP mining drill, slice, statistics and other operations can significantly improve the efficiency of multi-dimensional association rules. This paper studies multi-dimensional association rules with the aspect of the use of the Mondrian OLAP engine. Mondrian is implemented in Java language interpreted MDX parser. The combination between Mondrian OLAP engine and presentation layer tools JPivot is widely used in the field of OLAP Server architecture. Because it can be conveniently integrated in various software tools, and in heterogeneous data warehouse environments, it supports a variety of back-end database. Therefore, it is as flexible implementation strategy for the multi-dimensional association rules to provide support to improve the efficiency of multi-dimensional association rules.In this paper, we find out multi-dimensional association rule mining algorithm (MApriori) based on Mondrian's platform combines the classic association rules (Apriori),which can effectively solve the traditional Apriori mining algorithm who runs slow, need to generate a large number of candidate sets, frequently scan data cube to calculate the support of the shortcomings. Proved by this paper, the MApriori algorithm based on the platform of Mondrian is effective and stable in multidimensional association rule mining and has better performance than the traditional Apriori algorithm. Finally, The MApriori algorithm based on the platform of Mondrian will be used in the field of highway traffic accidents, to dig out some of the main factors which cause the traffic accident. The practical value of the MApriori algorithm is further confirmed.