The Application Research Based On Multidimensional Association Rule Mining Algorithm

Data is a valuable resource in the new developing century.Exploring the useful knowledge by using data mining techniques to provide critical data support for decision makers.Association rule mining is one of the most important means of data mining,and it has a unique role in mining transaction database.The article mainly focuses on mining medical data based on multi-dimensional association rules.Pre-processing the data before the mining and then use the Apriori algorithm to mine the data.In the process of mining,it found that there are many different values in a property of medical data.When facing this data,the improvement of the original algorithm is needed so that it can find the rules more quickly.By uniquely encoding the attribute and attribute values,it is determined whether the attribute value continues to be connected if the attribute of the new connection value already exists in the original frequent item set when each new attribute is connected.In view of the contents of this research,the main research points of this paper are:1.Using multi-dimensional association rules to mine medical data.It found that the impact of a single drug on the human body,also it provided data support for effective drug management and storage,according to the use of drug doses.2.An improved Apriori algorithm based on multi-valued attributes is proposed.The algorithm can effectively avoid the same attribute value connected when mining the medical data.The article did a detailed elaboration on the use of multi-dimensional association rules and Apriori algorithm innovation.By mining medical data,the article found that a single drug impact on certain indicators of the human body,and it can help hospitals and related personnel to patients with reasonable medication.The improved Apriori algorithm is used to mine the multi-valued attribute association rules.And it is found that the improved algorithm has a significant improvement in mining efficiency,by contrasting the Apriori algorithm and FP-tree algorithm.