关联规则的并行挖掘模型研究

Research on Parallel Mining Model of Association Rules

针对单机数据挖掘算法的现状,本文详细分析了Agrawal关联规则的并行挖掘模型,提出了基于客户机/服务器的并行挖掘模型和算法,建立了原型系统,该模型可以有效地用在关联规则的并行挖掘过程中,它能在一定程度上克服数据传输量大,无法及时得到所需数据而导致的效率不高的问题。

With the present situation of one machine data mining algorithms, the parallel mining model of Agrawal association rules is analyzed detailedly in this paper, a client/server based parallel mining model and algorithm are proposed, and a prototype system is created in the meantime. This model can be efficiently used in the process of parallel mining association rules, it can partly overcome large data transmission and low efficient problem caused by not receiving the needed data in time.