Research On The Algorithm Of Mining Association Rules In Data Mining

Association rules mining is an essential research topic in data mining. Presenting association rules mining initially aims at discovering interesting relation between different items in great deal of data. This paper study the data mining technique and association rules mining theory and algorithm based on the current research results; then present an improved algorithm based on the Apriori algorithm and analyze its capability; discuss correlated association rules mining algorithm.Firstly , reviews the definition, method , function, sort and research direction of Data mining. Then, introduces the basic concepts,theory and sorts of association rules mining ,discuss a series of improved algorithms, and point out the existent problem of the association rules mining algorithm which based on the support-confidence framework.Secondly, analyze and discuss the classic association rules mining algorithm-Apriori algorithm,and point out its theoretical value and performance bottlenecks . Based on the in-depth analysis and research for the complexity and the results problem of association rules mining algorithm, present an improved algorithm and argue its thinking; and the validity of this improved method is analyzed and testified by an example.Finally, analyze the correlation of association rules for the existing problems of the association rules mining algorithms which based on the support - confidence framework ,then discuss several existing correlated association rules mining algorithms,and present the optimum strategy for their limitations .Association rules mining is widely applied in various fields. The discovery,comprehension and application of association rules are important means of accomplishing the task of data mining.Therefore,the research of association rules mining is of great importance in both theoretical field and realistic field.