基于支持向量回归的多时间序列自回归方法

doi:10.3724/SP.J.1087.2012.02508

计算机应用 ›› 2012, Vol. 32 ›› Issue (09): 2508-2511.DOI: 10.3724/SP.J.1087.2012.02508

基于支持向量回归的多时间序列自回归方法

张伟¹,柳先辉^1*,丁毅²,史德明²

1.同济大学电子与信息工程学院,上海 201804;
2.马鞍山钢铁股份有限公司能源与环境保护部,安徽马鞍山 243000

收稿日期:2012-03-02 修回日期:2012-05-02 发布日期:2012-09-01 出版日期:2012-09-01
通讯作者: 柳先辉
作者简介:张伟(1985-),男,江西瑞昌人,博士研究生,主要研究方向:机器学习、数据挖掘; 柳先辉(1979-),男,浙江丽水人,讲师,博士研究生,主要研究方向:数据挖掘、企业信息化、软件工程。
基金资助:
国家863计划项目(2009AA043503);国家科技支撑计划项目(2012BAF10B05)

Multiple time series autoregressive method based on support vector regression

ZHANG Wei¹,LIU Xian-hui^1*,DING Yi²,SHI De-ming²

1.College of Electronics and Information Engineering,Tongji University,Shanghai 201804,China;
2.Department of Energy and Environmental Protection,Maanshan Iron and Steel Company Limited,Maanshan Anhui 243000,China

Received:2012-03-02 Revised:2012-05-02 Online:2012-09-01 Published:2012-09-01

摘要/Abstract

摘要： 能耗时间序列涉及多种能源,且各种能源间关系复杂,主要通过多个独立的单时间序列进行预报,这种方式忽略了多时间序列之间的依赖性。为了充分利用多时间序列之间的关联信息以提高预报的准确性,根据机器学习中的向量值函数学习和多任务学习理论,采用支持向量回归(SVR)算法建立了多时间序列的向量值自回归方法和多任务自回归方法。实验结果证明,与多个独立的单时间序列模型相比,通过这种方法建立的多时间序列自回归模型在焦化工序能耗预报中表现出了更好的性能。

关键词: 能耗, 多时间序列, 向量值函数学习, 多任务学习, 自回归方法, 支持向量回归

Abstract: Energy consumption time series involves a variety of energy and the relationship between different energy is complicated. Most existing consumption methods make prediction through multiple independent single time series respectively, which ignores dependencies between multiple time series. In order to take full advantage of the association between multiple time series and improve prediction accuracy, the vector-valued autoregressive method and multi-task autoregressive method based on Support Vector Regression (SVR) machines were proposed for multiple time series forecast according to vector-valued function learning and multi-task learning theory. The experimental results with energy consumption of coking process verify that multiple time series autoregressive models based on the proposed methods show better prediction performance.

Key words: energy consumption, multiple time series, vector-valued function learning, multi-task learning, autoregressive method, Support Vector Regression (SVR)

中图分类号:

TP181

张伟柳先辉丁毅史德明. 基于支持向量回归的多时间序列自回归方法[J]. 计算机应用, 2012, 32(09): 2508-2511.

ZHANG Wei LIU Xian-hui DING Yi SHI De-ming. Multiple time series autoregressive method based on support vector regression[J]. Journal of Computer Applications, 2012, 32(09): 2508-2511.

参考文献

[1]戴龙基,蔡蓉华.时间序列分析预测与控制[M].北京:北京大学出版社,2004.
[2]HELMUT L.New introduction to multiple time series analysis [M].Berlin:Springer,2005.
[3]PHILLIPS P C B,DURLAUF S N.Multiple time series regression with integrated processes [J].Review of Economic Studies,1986,53(4):473-495.
[4]ROSIPAL R,TREJO L J.Kernel partial least squares regression in reproducing kernel Hilbert space [J].Journal of Machine Learning Research,2001,2(12):97-123.
[5]MICCHELLI C A,PONTIL M.On learning vector-valued functions [J].Neural Computation,2005,17(1):177-204.
[6]CONTI S,O'HAGAN A.Bayesian emulation of complex multi-output and dynamic computer models [J].Journal of Statistical Planning and Inference,2010,140(3):640-651.
[7]PREZ-CRUZ F,CAMPS-VALLS G,SORIA-OLIVAS E,et al.Multi-dimensional function approximation and regression estimation [C]// Proceedings of the 2002 International Conference of Artificial Neural Networks.Berlin:Springer-Verlag,2002:757-762.
[8]EVGENIOUS T,PONTIL M.Regularized multi-task learning [C]// Proceedings of the 2004 International Conference on Knowledge Discovery and Data Mining.New York:ACM Press,2004:109-117.
[9]DEBASISH B,SRIMANTA P,DIPAK C P.Support vector regression [J].Neural Information Processing,2007,11(10):203-224.
[10]MARK B.Vector-valued support vector regression[C]// Proceedings of the 2006 International Joint Conference on Neural Networks.Piscataway,NJ:IEEE Press,2006:1562-1569.
[11]VAZQUEZ E,WALTER E.Multi-output support vector regression[C]// Proceedings of the 13th IFAC Symposium on System Identification.Piscataway,NJ:IEEE Press,2003:1820-1825.
[12]HASTIE T,TIBSHIRANI R,FRIEDMAN J.The elements of statistical Learning [M].2nd ed.Berlin:Springer-Verlag,2009.
[13]CAO L J,TAY E H.Support vector machine with adaptive parameters in financial time series forecasting [J].IEEE Transactions on Neural Networks,2003,11(6):1506-1518.
[14]KIM K.Financial time series forecasting using support vector machines [J].Neurocomputing,2003,55(2003):307-319.
[15]CARMELI C,VITO E D,TOIGO A.Vector valued reproducing kernel Hilbert spaces of integrable functions and Mercer theorem [J].Analysis and Applications,2006,4(4):377-408.
[16]EVGENIOU T,MICCHELLI C A,PONTIL M.Learning multiple tasks with kernel methods [J].Journal of Machine Learning Research,2005,6(4):615-637.

基于支持向量回归的多时间序列自回归方法

Multiple time series autoregressive method based on support vector regression

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	武国亮, 徐继宁. 基于命名实体识别任务反馈增强的中文突发事件抽取方法[J]. 计算机应用, 2021, 41(7): 1891-1896.
[2]	姚杰, 程春玲, 韩静, 刘峥. 云工作流中基于多任务时序卷积网络的异常检测方法[J]. 计算机应用, 2021, 41(6): 1701-1708.
[3]	章荪, 尹春勇. 基于多任务学习的时序多模态情感分析模型[J]. 计算机应用, 2021, 41(6): 1631-1639.
[4]	杨翎, 姜春茂. 基于三支决策的虚拟机节能迁移策略[J]. 计算机应用, 2021, 41(4): 990-998.
[5]	孙环, 陈宏滨. 基于萤火虫算法的无线传感器网络节点重部署策略[J]. 计算机应用, 2021, 41(2): 492-497.
[6]	阮灿华, 林甲祥. 动态事件时间数据的多任务Logistic生存预测方法[J]. 计算机应用, 2020, 40(5): 1284-1290.
[7]	赵志学, 李夏苗, 周鲜成. 考虑拥堵区域的多车型绿色车辆路径问题优化[J]. 计算机应用, 2020, 40(3): 883-890.
[8]	潘建国, 李豪. 基于实用拜占庭容错的物联网入侵检测方法[J]. 计算机应用, 2019, 39(6): 1742-1746.
[9]	李启锐, 彭志平, 崔得龙, 何杰光. 容器云环境虚拟资源配置策略的优化[J]. 计算机应用, 2019, 39(3): 784-789.
[10]	曹金梦, 倪蓉蓉, 杨彪. 基于多尺度多任务卷积神经网络的人群计数[J]. 计算机应用, 2019, 39(1): 199-204.
[11]	李树, 于炯, 国冰磊, 蒲勇霖, 杨德先, 刘粟. 基于SSD数据库负载的SQL能耗感知模型[J]. 计算机应用, 2019, 39(1): 205-212.
[12]	汪汉新, 洪思琴. 基于环分块的能耗均衡分簇路由算法[J]. 计算机应用, 2019, 39(1): 251-255.
[13]	毛莺池, 郝帅, 平萍, 戚荣志. 基于组合双向拍卖的云资源调度方法[J]. 计算机应用, 2019, 39(1): 1-7.
[14]	李安超, 陈桂芬. 基于奇偶轮成簇和双簇首的非均匀分簇协议[J]. 计算机应用, 2018, 38(7): 1995-2000.
[15]	佟飘, 龙隆, 韩雪, 邱大伟, 胡茜. 基于簇内能耗最优的设备到设备缓存通信内容共享算法[J]. 计算机应用, 2018, 38(6): 1703-1708.