Reducing and mining of frequent patterns in distributed database
Han Jiang · Journal of Zhejiang Normal University · 2010
Based on thorough investigation of mining and simplifying association rules on distributed databases,the concept of closed frequent itemsets and maximal frequent itemsets was extended to combining with distributed databases.Furthermore,it was proposed a new simplification technique of frequent patterns in distributed environment,together with a fresh mining algorithm of globally closed frequent patterns itemsets and globally maximal frequent itemsets,which was based on local-global itemsets at each distributed station.The characteristics of the proposed methods costed less memory and with less traffic load.The experiments demonstrated the competence of the new algorithms.