Mining Association Rules Based On Apriori Algorithm And Improvement

Mining association rules is an important topic in the data mining research. Apriori algorithm submitted by Agrawal and R. Srikant in 1994 is the most effective algorithm. In this paper, I present recognizable matrix, recognizable vector, interestingness, association rules including negative items and submit an improved Apriori algorithm, aim at low efficiency of mining frequent itemsets and creating useless rules, losing useful rules. Based on my new method, mining frequent itemsets is more effective and creating association rules is more reasonable.This paper is mainly composed of the following result:1. Putting forward mining association rules improved algorithm based on recognizable matrix.2. Defining interestingness and putting forward new algorithm of mining association rules including negative items.3. The above method is used to mine the students' datatable under Visual Foxpro 6. 0.