The Research & Application On Association Rule Mining Algorithm

The high speed development of the Information technology and Internet network'spopularization,caused in the database in many application domains to save the magnanimous data,and people urgently want to convert these datas to the information and the knowledge.Because of this,the data mining technology arises;this technology devotes in gaining the interior implication rich knowledge from the mass data.The association rules mining is a important research branch in data mining.At present it already had the widespread application.Apriori algorithm is the classical algorithm in the association rule mining,when it mines the big database,its efficiency is very low;Based on this kind of situation,there are important theory value and practical significance in association rules mining algorithm research.The association rule mining has two steps,including mining the frequent item set and producing the rules,among of them,the key is the mining the frequent item set.Therefore,designing the highly effective mining algorithm about the frequent item set is an important research content.Based on the importance of the frequent item set mining algorithm,as welt as the insufficiency of the classics Apriori algorithm,this paper proposes the improvement of the Apriori algorithm:one is the transaction database mapping for an across linker description,the other is the transaction database mapping for an point linke description,which can reduce the time cost of the database linking and the scanning time of the transaction record.At last,by analyzing the efficiency of the Apriori algorithm and the improved algorithm,it shows the improved algorithm can enhance the execution effciency.