Design About Association Rules Mining Based On Items Clustering And Transaction Tree

Association rules mining, as the most important subject in data mining, reveals the correlations between itemsets and therefore can be widely applied to many fields such as market basket analysis, correlation analysis, classification, web customized service, etc. Since 1993 R. Agrawal, R. Srikant firstly proposed the concept of association rules, a lot of algorithms have come up in mining of association rules. While most of these are based on Apriori algorithm, will generate a huge number of candidate itemsets, need multiple scans over database, and maintain a big hash tree, so the time and space complexity is too high.This paper proposed an algorithm based on division--YD_Apriori. It scans the database only once and find frequent 1 itemsets, then divides database into several parts according to different item_supprort, after that use transaction tree to generate local frequent items, at last merge every linklist one by one according to transaction tree, and generate new frequent items, all the frequent items are what we want. The method of generating association rules is the same as method which Apriori generate, and we get the association rules we want. The aim of this paper is concentrated on how to generate frequent items.