摘要
关联规则是数据挖掘中的一个重要研究内容 .本文提出了 Shared- nothing并行事务数据库系统 (简称SNPDBS)中一种快速的关联规则挖掘算法 SNPMAR,并考虑当最小支持度发生变化后 SNPDBS中关联规则的高效更新问题 ,提出了一种有效的关联规则更新算法 SNPIUA.
Discovering association rules is an important data mining problem, a lot of algorithms for mining association rules have been proposed in a single transaction database system.However, many large databases are distributed in nature, so the development of algorithms for efficient mining of association rules in a shared nothing parallel transaction database system has its unique importance.In this paper, an efficient algorithm SNPMAR and its incremental updating algorithm SNPIUA are presented to discovery association rules in a shared nothing parallel transaction database system.The algorithm SNPIUA will make use of the previous mining result to cut down the cost of finding new rules in an updated database.Comparing with CD algorithm, the author also offers some experiments to show that the new algorithm is more efficient.
出处
《小型微型计算机系统》
CSCD
北大核心
2003年第8期1499-1502,共4页
Journal of Chinese Computer Systems
基金
国家自然科学基金 ( 79970 0 92 )资助
国家科技型中小型企业创新基金 ( 0 0 C2 62 13 2 110 14 )资助
