Efficient Mining Of Association Rules In Distributed Database System

Data mining is an emerging research topic in database and artificial intelligence fields. It has attracted a great deal of attention in the information industry in recent years. The major reason is due to the wide applications ranging from business management, production control and market analysis to engineering design and science exploration.This paper introduces the techniques of data mining, including its producing background , its application , its classification and its algorithms. Efficient mining algorithm of association rules is the emphasis of the paper. The typical algorithms of mining association rules are discussed.The generation of frequent itemsets is a key problem of mining association rules. This paper presents an algorithm based on Hash Tree for generation of frequent itemsets. The algorithm contains two parts, including creation algorithm and recursion algorithm. The performance of the algorithm based on Hash Tree is studied. Compared with the traditional algorithm, the experiment shows that the algorithm based on Hash Tree is efficient.The updating problem of association rules in distributed database system is researched in this paper also. An updating algorithm DUA for maintaining of association rules in distributed database system is presented. The algorithm DUA has simple time complexity and small communication cost. We design a system DAMINER for efficient mining association rules in distributed databases. The algorithm ApriorK DMA and DUA in DAMINER have been implemented using Java. The performance of DAMINER is studied. The results show that the algorithm DUA is valid and fast.