基于网格的并行数据挖掘算法的实现  

Realization of grid-based parallel data mining algorithm

作　　者：郑晶[1] 

机构地区：[1]福建经济管理干部学院,福建福州350001

出　　处：《福建工程学院学报》2010年第1期83-86,共4页Journal of Fujian University of Technology

基　　金：福建省教育厅A类科技项目(JA09272)

摘　　要：随着信息技术的发展,大量的数据不断被收集和存储,对数据的挖掘规模越来越大,传统的数据挖掘已经无法解决海量数据挖掘问题。网格技术的发展,使得广域分布的海量数据的挖掘问题得到解决。文章根据现实中海量数据挖掘的需求和网格的本质,把并行关联规则挖掘算法应用于网格数据挖掘中,并建立了一个并行数据挖掘模型,验证了并行数据挖掘系统在网格环境的可行性与合理性。The scale of data mining increases with the collection and storage of immense data resulting from the development of information technology. As the traditional way of data mining could not resolve massive data mining, grid technology for mining widely-distributed large amount of data was developed. Parallel association rules of mining algorithm were applied to grid data mining. A parallel data mining service model was constructed based on the requirement of numerous data mining and the nature of grid. The feasibility and rationality of the application of parallel data mining to the gird environment was confirmed.

关 键 词：网格 数据挖掘 并行算法 

分 类 号：TP393.2[自动化与计算机技术—计算机应用技术]