基于云计算及数据挖掘技术的海量数据处理研究  

Research on Massive Data Processing Based on Cloud Computing and Data Mining

作　　者：王鹏[1] 王健安[1] 郭畅[1] 巴济慈 

机构地区：[1]长春理工大学计算机科学技术学院,长春130022

出　　处：《长春理工大学学报（自然科学版）》2013年第6期157-160,共4页Journal of Changchun University of Science and Technology(Natural Science Edition)

摘　　要：主要介绍了在HADOOP平台下使用优化的SPRINT算法进行海量数据处理。首先介绍了传统的数据挖掘算法SPRINT算法,然后结合云计算中的MapReduce编程模式对SPRINT算法进行改进和优化,将并行的SPRINT算法移植到HADOOP平台下,最后通过实验实现分布式数据处理。In this paper, optimized SPRINT algorithm which was used in the Hadoop platform for mass data process- ing was introduced. Firstly, the SPRINT algorithm which is the traditional data mining algorithm was introduced and then was combined with the Map Reduce model in the cloud computing to improve and optimize the SPRINT algo- rithm. At last, the parallel SPRINT algorithm will be transplanted to the HADOOP platform, and finally the distribut- ed data processing was achieved through the experiment.

关 键 词：云计算 数据挖掘 SPRINT HADOOP MAPREDUCE 

分 类 号：TP31[自动化与计算机技术—计算机软件与理论]