| 一种基于MapReduce的粗糙集并行属性约简算法 |
| |
| 引用本文： | 杨勇,朱影.一种基于MapReduce的粗糙集并行属性约简算法[J].重庆邮电大学学报(自然科学版),2015,27(1):89-96. |
| |
| 作者姓名： | 杨勇 朱影 |
| |
| 作者单位： | 重庆邮电大学计算智能重庆市重点实验室,重庆,400065 |
| |
| 基金项目： | 重庆市自然科学基金(CSTC,2007BB2445);重庆市教委科学技术研究项目(KJ110522);重庆邮电大学科研基金(A2009-26) |
| |
| 摘 要： | 云计算技术是海量数据挖掘的一种高效解决方案,将MapReduce并行计算模型与粗糙集属性约简算法相结合,提出一种基于MapReduce的浓缩布尔矩阵并行属性约简算法.该算法提高了粗糙集属性约简算法对大数据的处理能力和效率,并能适应云计算环境.实验结果表明,所提算法具有良好的效率、加速比和可扩展性.
|
| 关 键 词： | MapReduce 粗糙集 属性约简 浓缩布尔矩阵 |
| 收稿时间： | 2014/2/12 0:00:00 |
| 修稿时间： | 2014/10/23 0:00:00 |
A parallel rough set attribute reduction algorithm based on MapReduce |
| |
| YANG Yon and ZHU Ying.A parallel rough set attribute reduction algorithm based on MapReduce[J].Journal of Chongqing University of Posts and Telecommunications,2015,27(1):89-96. |
| |
| Authors: | YANG Yon and ZHU Ying |
| |
| Institution: | Institute of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, P. R. China and Institute of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, P. R. China |
| |
| Abstract: | Cloud computing technology is a high efficient solution for massive data mining. MapReduce, the parallel computing model is combined with attribute reduction algorithm of rough set, and a novel parallel concentration Boolean matrix attribute reduction algorithm is proposed in this paper. The algorithm improves the capacity and efficiency of the rough set attribute reduction algorithms for big data. Furthermore, it also adapts to the cloud computing environment. The experimental results illustrate the high efficiency, speedup and scaleup of the proposed algorithm. |
| |
| Keywords: | MapReduce rough set attribute reduction concentration Boolean matrix |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《重庆邮电大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《重庆邮电大学学报(自然科学版)》下载免费的PDF全文 |
