云计算下的海量数据挖掘研究

Research on Mass Data Mining under Cloud Computing

云计算的出现为愈来愈多的中小企业分析海量数据提供廉价的解决方案。在介绍基于云计算的Hadoop集群框架和数据挖掘技术中的SPRINT分类算法的基础上,详细描述SPRINT并行算法在Hadoop中的MapReduce编程模型上的执行流程,并利用分析出的决策树模型对输入数据进行分类。

Cloud Computing provides a low-priced way for small and medium sized enterprises to analyze mass data. Based on Hadoop of Cloud Computing and SPRINT algorithm of data mining, proposes the detailed procedure of SPRINT algorithm on MapR.eduee, and classifies the input data is by the model of decision tree.