基于稀疏贝叶斯学习的协同进化时间序列缺失数据预测算法

doi:10.11896/j.issn.1002-137X.2019.07.033

计算机科学 ›› 2019, Vol. 46 ›› Issue (7): 217-223.doi: 10.11896/j.issn.1002-137X.2019.07.033

基于稀疏贝叶斯学习的协同进化时间序列缺失数据预测算法

宋晓祥,郭艳,李宁,余东平

(陆军工程大学通信工程学院南京210007)

收稿日期:2018-05-27 出版日期:2019-07-15 发布日期:2019-07-15
作者简介:宋晓祥(1993-),男,硕士生,主要研究方向为信号处理、大数据;郭艳(1971-),女,博士,教授,博士生导师,主要研究方向为波束形成、认知无线电、无线传感器网络定位、自适应信号处理,E-mail:guoyan_1029@sina.com(通信作者);李宁(1967-),男,副教授,硕士生导师,主要研究方向为Ad hoc网络、无线认知网络;余东平(1989-),男,博士,主要研究方向为无线传感器网络定位、信号处理。
基金资助:
国家自然科学基金(61571463,61371124,61472445),江苏省自然科学基金(BK20171401)资助

Missing Data Prediction Algorithm Based on Sparse Bayesian Learning in Coevolving Time Series

SONG Xiao-xiang,GUO Yan,LI Ning,YU Dong-ping

(College of Communications Engineering,Army Engineering University of PLA,Nanjing 210007,China)

Received:2018-05-27 Online:2019-07-15 Published:2019-07-15

摘要/Abstract

摘要： 针对大多数已有算法在预测协同进化时间序列中的缺失数据时只适用于缺失数据较少情况的问题,提出了一种高效的缺失数据预测算法。首先,应用压缩感知理论,将协同进化时间序列中的缺失数据预测问题建模成多稀疏向量恢复问题;其次,从稀疏表示向量是否足够稀疏和感知矩阵是否满足有限等距特性两个方面分析了模型的性能;最后,针对协同进化时间序列的特点设计了一种基于稀疏贝叶斯学习的高效恢复算法,该算法可以通过学习得到部分支持信息,从而同时解决多个稀疏向量的恢复问题。仿真结果表明,所提算法可以同时有效地预测出多个时间序列中的缺失数据。

关键词: 感知矩阵, 缺失数据, 稀疏贝叶斯学习, 稀疏表示向量, 协同进化时间序列

Abstract: In view of most of the existing algorithms in predicting the missing data in the coevolving time series are only feasible to be applied to the case where only a low ratio of collected data are missing,an efficient missing data prediction method was proposed in this paper.Firstly,the compressive sensing theory is applied to model the missing data prediction problem in the coevolving time series to the problem of multiple sparse vectors recovery.Secondly,the validity of the model is analyzed from two aspects:whether the sparse representation vector is sufficiently sparse and the sensing matrix satisfies the restricted isometry property.Finally,the novel recovery algorithm based on sparse Bayesian lear-ning,which can solve multiple sparse vector recovery problems by learning some support information,is designed for the characteristics of coevolving time series.Simulation results show that the proposed algorithm can effectively predict the missing data in multiple time series simultaneously.

Key words: Coevolving time series, Missing data, Sensing matrix, Sparse bayesian learning, Sparse representation vector

中图分类号:

TN911.7

宋晓祥,郭艳,李宁,余东平. 基于稀疏贝叶斯学习的协同进化时间序列缺失数据预测算法[J]. 计算机科学, 2019, 46(7): 217-223. https://doi.org/10.11896/j.issn.1002-137X.2019.07.033

SONG Xiao-xiang,GUO Yan,LI Ning,YU Dong-ping. Missing Data Prediction Algorithm Based on Sparse Bayesian Learning in Coevolving Time Series[J]. Computer Science, 2019, 46(7): 217-223. https://doi.org/10.11896/j.issn.1002-137X.2019.07.033

参考文献

[1]SOWMYA R,SUNEETHA K R.Data mining with big data[C]∥ International Conference on Intelligent Systems and Control.IEEE,2017:246-250.
[2]SHI W,ZHU Y,YU P S,et al.Temporal dynamic matrix factori- zation for missing data prediction in large scale coevolving time series [J].IEEE Access,2017,4(99):6719-6732.
[3]ELENI I,VLAHOGIANNI J,GOLIAS C,et al.Short-term traffic forecasting:Overview of objectives and methods [J].Transport Reviews,2004,24(5):533-557.
[4]BALOUJI E,SALOR Q,ERMIS M.Exponential smoothing of multiple reference frame components with GPUs for real-time detection of time-varying harmonics and inter harmonics of EAF currents[C]∥IEEE Industry Applications Society Meeting.IEEE,2017:1-8.
[5]KOZERA R,WILKOLAZKA M.Natural spline interpolation and exponential parameterization for length estimation of curves[C]∥International Conference of Numerical Analysis & Applied Mathematics.AIP Publishing LLC,2017:1-140.
[6]AL-DEEK H M,CHANDRA C.New algorithms for filtering and imputation of real-time and archived dual-loop detector data in 1-4 data warehouse[C]∥Meeting of the Transportation-Research-Board.2004:116-126.
[7]BOYLES S D.A comparison of interpolation methods for mis- sing traffic volume data[C]∥Transportation Research Board Annual Meeting.2011:23-27.
[8]LIPPI M,BERTINI M,FRASCONI P.Short-term traffic flow forecasting:An experimental comparison of time-series analysis and supervised learning[J].IEEE Transactions on Intelligent Transportation Systems,2013,14(2):872-882.
[9]LI Y,LI Z,LI L,et al.Comparison on PPCA,KPPCA and MPPCA based missing data imputing for traffic flow[C]∥Procee-dings of IEEE Conference on Intelligent Transportation System.2013:1535-1540.
[10]SHI W,ZHU Y,YU P.Effective Prediction of Missing Data on Apache Spark over Multivariable Time Series[J].IEEE Tran-sactions on Big Data,2017,PP(99):1.
[11]STRAUMAN A S,BIANCHI F M,MIKALSEN K.Classification of postoperative surgical site infections from blood measurements with missing data using recurrent neural networks[C]∥IEEE EMBS International Conference on Biomedical & Health Informatics (BHI).2018:307-310.
[12]CAI Y,TONG H,FAN W,et al.Fast mining of a network of coevolving time series[C]∥2015 SIAM International Conference on Data Mining.2015:298-306.
[13]HADI A,WAHIDAH I.Delay estimation using compressive sensing on WSN IEEE 802.15.4[C]∥International Conference on Control,Electronics,Renewable Energy and Communications.IEEE,2017:192-197.
[14]ARJOUNE Y,KAABOUCH N,GHAZI H E,et al.Compressive sensing:Performance comparison of sparse recovery algorithms[C]∥Computing and Communication Workshop and Con-ference.IEEE,2017:1-7.
[15]RHEE I,SHI N M,HONG S,et al.Mobility traces[OL]. http://carwdad.org/ncsu/mobilitymodels.
[16]SAMUEL M.Intel Lab Data[OL].http://db.csail.mit.edu.
[17]FONOLLOSA J,SHEIK S,HUERTA R,et al.Reservoir computing compensates slow response of chemo sensor arrays exposed to fast varying gas concentrations in continuous monitoring[J].Sensors,2015,215:618-629.
[18]WU X,LIU M.In-situ soil moisture sensing:Measurement scheduling and estimation using compressive sensing[C]∥Proceedings of the 11th ACM International Conference on Information Processing in Sensor Networks.2012:1-12.
[19]HU S W,LIN G X,HSIEH S H,et al.Performance analysis of joint-sparse recovery from multiple measurement vectors with prior information via convex optimization[C]∥IEEE International Conference on Acoustics,Speech and Signal Processing.IEEE,2016:4368-4372.
[20]ZHAO X,YANG Q,ZHANG Y.Synthesis of sparse linear array with multiple patterns based on joint sparse recovery[C]∥IEEE International Symposium on Antennas and Propagation & Usnc/ursi National Radio Science Meeting.IEEE,2017:425-426.
[21]WALEWSKI A C,STEFFENS C,PESAVENTO M.Off-Grid Parameter Estimation Based on Joint Sparse Regularization[C]∥International Itg Conference on Systems,Communications and Coding.2017.
[22]ZHANG Z,RAO B D.Sparse signal recovery in the presence of correlated multiple measurement vectors[C]∥International Conference on Acoustics Speech and Signal Processing.2010:3986-3989.
[23]PRASAD R,MURPHY C R,RAO B D.Joint approximately sparse channel estimation and data detection in OFDM systems using sparse Bayesian learning[J].IEEE Signal Processing Letters,2014,62(14):3591-3603.
[24]CHEN W.Simultaneous sparse Bayesian learning with partially shared support[J].IEEE Signal Processing Letters,2017,24(11):1641-1645.
[25]TIPPING M E.Sparse Bayesian learning and the relevance vector machine[J].Journal of Machine Learning Research,2001,1(3):211-244.
[26]BISHOP C M.Pattern Recognition and Machine Learning (Information Science and Statistics).Springer-Verlag New York,2006.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于稀疏贝叶斯学习的协同进化时间序列缺失数据预测算法

Missing Data Prediction Algorithm Based on Sparse Bayesian Learning in Coevolving Time Series

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

Metrics

本文评价

推荐阅读 0

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

[1]	李培冠, 於志勇, 黄昉菀. 基于稀疏表示的电力负荷数据补全 Power Load Data Completion Based on Sparse Representation 计算机科学, 2021, 48(2): 128-133. https://doi.org/10.11896/jsjkx.191200152
[2]	陈平, 郭秋歌, 李攀, 崔峰. OFDM系统中基于贝叶斯学习的联合稀疏信道估计与数据检测 Joint Sparse Channel Estimation and Data Detection Based on Bayesian Learning in OFDM System 计算机科学, 2020, 47(11A): 349-353. https://doi.org/10.11896/jsjkx.191100090
[3]	宋晓祥, 郭艳, 李宁, 王萌. 基于压缩感知的时间序列缺失数据预测算法 Missing Data Prediction Based on Compressive Sensing in Time Series 计算机科学, 2019, 46(6): 35-40. https://doi.org/10.11896/j.issn.1002-137X.2019.06.004
[4]	卞孝丽. 基于拉伸因子图的低复杂度贝叶斯稀疏信号算法研究 Low Complexity Bayesian Sparse Signal Algorithm Based on Stretched Factor Graph 计算机科学, 2018, 45(6A): 135-139.
[5]	陈静杰,车洁. 基于标准欧氏距离的燃油流量缺失数据填补算法 Fuel Flow Missing-value Imputation Method Based on Standardized Euclidean Distance 计算机科学, 2017, 44(Z6): 109-111. https://doi.org/10.11896/j.issn.1002-137X.2017.6A.023
[6]	王锋魏巍. 缺失数据数据集的组增量式特征选择 Group Feature Selection Algorithm for Data Sets with Missing Data 计算机科学, 2015, 42(7): 285-290. https://doi.org/10.11896/j.issn.1002-137X.2015.07.061
[7]	彭红毅朱思铭蒋春福. 数据挖掘中基于ICA的缺失数据值的估计计算机科学, 2005, 32(12): 203-205.