《计算机应用研究》|Application Research of Computers

基于MapReduce的最小二乘支持向量机回归模型

Least squares support vector machine regression model based on MapReduce

免费全文下载 (已被下载 次)  
获取PDF全文
作者 代亮,许宏科,陈婷,钱超,梁殿鹏
机构 1.长安大学 a.电子与控制工程学院;b.信息工程学院,西安 710064;2.IBM中国系统与科技开发中心,西安 710068
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2015)04-1060-05
DOI 10.3969/j.issn.1001-3695.2015.04.024
摘要 针对最小二乘支持向量机处理大规模数据集耗时长且受内存限制的特点,将局部多模型方法与Map-Reduce编程模式相结合,提出一种并行最小二乘支持向量机回归模型。模型由两组MapReduce过程组成,首先按照输入样本集对样本数据进行聚类操作,再对聚类后得到的子类按输出样本集进行二次聚类操作,分别得到局部模型数目和各局部模型综合加权输出计算结果。实验结果表明,并行最小二乘支持向量机回归模型具有较好的加速比和可扩展性。
关键词 最小二乘支持向量机;MapReduce编程模式;局部多模型方法;加速比;可扩展性
基金项目 国家自然科学基金资助项目(51308057,51378073)
中国博士后科学基金面上资助项目(2014M550475)
国家教育部创新团队发展计划资助项目(IRT1050)
交通运输部基础研究基金资助项目(2010-319-812-080)
陕西省自然科学基础研究计划资助项目(2014JQ8354)
中央高校基本科研业务费专项资金资助项目(0009-2014G1321041,2013G3324005)
本文URL http://www.arocmag.com/article/01-2015-04-024.html
英文标题 Least squares support vector machine regression model based on MapReduce
作者英文名 DAI Liang, XU Hong-ke, CHEN Ting, QIAN Chao, LIANG Dian-peng
机构英文名 1. a. School of Electronic & Control Engineering, b. School of Information Engineering, Chang'an University, Xi'an 710064, China; 2. IBM China Systems & Technology Laboratory, Xi'an 710068, China
英文摘要 According to the characteristics of least squares support vector machine regression model for long processing time and memory constraints, this paper designed a parallel least squares support vector machine regression model based on MapReduce and local multi-model method. The model was composed of two MapReduce process. It clustered the sample data accor-ding to the input set, and then obtained second clustering after sub set according to the output. Two MapReduce processes were calculated the number of local model and weighted output of each model. Experimental results show that the proposed parallel least squares support vector machine regression model has better speedup and scaleup.
英文关键词 least squares support vector machine; MapReduce programming pattern; local multi-model method; speedup; scaleup
参考文献 查看稿件参考文献
  [1] MIRANIAN A, ABDOLLAHZADE M. Developing a local least-squares support vector machines-based neuro-fuzzy model for nonlinear and chaotic time series prediction[J] . IEEE Trans on Neural Networks and Learning Systems, 2013, 24(2):207-218.
[2] WANG Li-guo, LIU Dan-feng, WANG Qun-ming, et al. Spectral unmixing model based on least squares support vector machine with unmixing residue constraints[J] . IEEE Geoscience and Remote Sensing Letters, 2013, 10(6):1592-1596.
[3] 周家中, 张殿业. 基于空间加权的LS-SVM城市轨道交通车站客流量预测[J] . 铁道学报, 2014, 36(1):1-7.
[4] 赵亚萍, 张和生, 周卓楠, 等. 基于最小二乘支持向量机的交通流量预测模型[J] . 北京交通大学学报, 2011, 35(2):114-117.
[5] LIU Xing-wei, FANG Xu-ming, QIN Zhen-hua, et al. A short-term forecasting algorithm for network traffic based on chaos theory and SVM[J] . Journal of Network and Systems Management, 2011, 19(4):427-447. [6] ZHANG Yang, LIU Yun-cai. Data imputation using least squares support vector machines in urban arterial streets[J] . IEEE Signal Processing Letters, 2009, 16(5):414-417.
[7] ZHANG Yang. Hourly traffic forecasts using interacting multiple mo-del (IMM) predictor[J] . IEEE Signal Processing Letters, 2011, 18(10):607-610.
[8] CHRISTIAN K, ALEXEI P. Enabling real-time city sensing with kernel stream oracles and MapReduce[J] . Pervasive and Mobile Computing, 2013, 9(5):708-721.
[9] 张浩然, 汪晓东. 回归最小二乘支持向量机的增量和在线式学习算法[J] . 计算机学报, 2006, 29(3):400-406.
[10] 叶洪涛, 罗飞. 基于免疫优化多输出最小二乘支持向量机及其应用[J] . 计算机应用研究, 2010, 27(6):2065-2067.
[11] 杨福刚. 基于人工免疫算法的最小二乘支持向量机参数优化算法[J] . 计算机应用研究, 2010, 27(5):1702-1704.
[12] 李建江, 崔健, 王聃, 等. MapReduce并行编程模型研究综述[J] . 电子学报, 2012, 39(11):2635-2642.
[13] 张永, 尹传晔, 吴崇正. 基于MapReduce的PageRank算法优化研究[J] . 计算机应用研究, 2014, 31(2):431-434.
[14] 唐多余, 曹菡. 基于MapReduce的加权Voronoi图并行算法设计及应用[J] . 计算机应用研究, 2013, 30(5):1410-1412, 1418.
[15] 郑莉华, 曾雪. 基于MapReduce的H. 264/AVC并行视频编码[J] . 计算机应用研究, 2013, 30(10):3139-3141, 3145.
[16] GUO Qi, LI Yan, LIU Tao, et al. Correlation-based performance analysis for full-system MapReduce optimization[C] //Proc of IEEE International Conference on Big Data. Washington DC:IEEE Compu-ter Society, 2013:753-761.
[17] BORIS S K. Introduction to modern traffic flow theory and control[M] . Berlin:Springer-Verlag, 2009.
收稿日期 2014/3/24
修回日期 2014/5/16
页码 1060-1064
中图分类号 TP181
文献标志码 A