Research And Application Of Frequent Itemsets Mining Algorithm

With the development of Internet technology,data from all walks of life have shown explosive growth.In order to be able to extract useful information from a huge database,people have more and more research on data mining technology.Association rule mining technology is one of the main research contents of data mining.The association rule mining algorithm is used to mine frequent itemsets in transaction data,and then find the association rules that satisfy certain conditions from these frequent itemsets.Interesting relevance,which in turn provides decision makers with scientific decision-making and technical support.However,the classic algorithm is based on the premise of exchanging frequent itemsets with the same importance.This is impractical in many industries because there are important differences between different items.It is impossible to distinguish their importance,so the mining results can not provide effective reference for decision makers.At the same time,the classic algorithm does not consider the reality that transaction data and threshold are dynamic changes when mining frequent itemsets.Analysis of static data alone is likely to cause decision makers to make erroneous judgments about trends.On the other hand,when thresholds are adjusted,if the entire transaction database is re-excavated,it will consume a lot of time again.Based on these problems,this paper proposes the WDUA algorithm,which only scans the transaction database once,transforms the generation process of frequent itemsets into the vector operation process in the weighted association matrix of the item set,and then better excavates the data through the custom tree structure DCTree.The intrinsic connection of the weighted frequent itemsets and the mining process of the new frequency set,combined with the constructed auxiliary matrix and the tree structure,can still efficiently mine frequent itemsets when data updates or threshold changes.Through a large number of experiments and comparison with classical mining algorithms,it shows that the frequent itemsets mined by theimproved algorithm have higher correctness,and the execution efficiency is higher than the classical mining algorithm.