基于Hadoop的C4.5决策树分类算法并行化  

The parallelization of C4.5 algorithm based on Hadoop

作　　者：林树地[1] 吴扬扬[1] 

机构地区：[1]华侨大学计算机科学与技术学院,福建厦门361000

出　　处：《微型机与应用》2013年第12期85-87,91,共4页Microcomputer & Its Applications

基　　金：福建省科技计划重点项目(2011H0028)

摘　　要：通过研究各种决策树分类算法的并行方案后,并行设计C4.5算法。同时根据Hadoop云平台的MapReduce编程模型,详细描述C4.5并行算法在MapReduce编程模型下的实现及其执行流程。最后,对输入的海量文本数据进行分类,验证了算法的高效性和扩展性。In this paper, a parallel C4.5 algorithm is put forward by the study of a variety of decision tree classification algo rithm parallel programs and the MapReduce programming model of the Hadoop cloud computing platform. At the same time, the ex ecution flow of the C4.5 parallel algorithms in the MapReduce programming model is introduced. Finally, the input of mass text da ta is classificated to verify the efficiency and scalability of the algorithm.

关 键 词：云计算 HADOOP MAPREDUCE 数据分类 C4 5算法 并行 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]