基于黎曼流形的多元时间序列异常检测
详细信息 本馆镜像全文    |  推荐本文 | | 获取馆网全文
摘要
多元时间序列问题广泛存在于社会生产和生活中,异常检测已经在金融、水文、气象、地震、视频监控医疗以及其他领域给人们提供了很多有价值的信息。为了快速高效地发现时间序列中的异常,使之以直观的方式呈现在人们面前,本文以滑动窗口为基础,用协方差矩阵作为时间序列的描述子,将黎曼流形与统计过程控制图相结合,来实现多元时间序列的异常检测及其可视化。以MA模拟数据流和MIT-BIH的心电失常数据作为实验对象,对异常检测方法进行了验证,结果表明该方法是合理有效的。
Multivariate time series problems widely exist in production and life in the society.Anomaly detection has provided people with a lot of valuable information in financial,hydrological,meteorological fields,and the research areas of earthquake,video surveillance,medicine and others.In order to quickly and efficiently find exceptions in time sequence so that it can be presented in front of people in an intuitive way,we in this study combined the Riemannian manifold with statistical process control charts,based on sliding window,with a description of the covariance matrix as the time sequence,to achieve the multivariate time series of anomaly detection and its visualization.We made MA analog data flow and abnormal electrocardiogram data from MIT-BIH as experimental objects,and verified the anomaly detection method.The results showed that the method was reasonable and effective.
引文
[1]CHANDOLA V,BANERJEE A,KUMAR V.Anomaly detection:a survey[J].ACM Computing Surveys(CSUR),2009,41(3):15.
    [2]POVINELLI R J,FENG X.A new temporal pattern identification method for characterization and prediction of complex time series events[J].IEEE Transactions on Knowledge and Data Engineering,2003,15(2):339-352.
    [3]PAPADIMITRIOU S,SUN J,FALOUTSOS C.Streaming pattern discovery in multiple time-series[C]//Proceedings of the 31st international conference on Very large data bases.VLDB Endowment,Trondheim:2005:697-708.
    [4]BREUNIG M M,KRIEGEL H P,NG R T,et al.LOF:identifying density-based local outliers[C]//ACM Sigmod Record.Dallas:2000:93-104.
    [5]王欣.两阶段的多元时间序列异常检测算法[J].计算机应用研究,2011,28(7):2466-2469.
    [6]李权,周兴社.一种新的多变量时间序列数据异常检测方法[J].时间频率学报,2011,34(2):154-158.
    [7]郝井华,刘民,吴澄,等.一种基于LLM的高维时间序列数据异常检测方法[J].控制工程,2005,12(3):207-209,265.
    [8]TOKUDA T,GOODRICH B,VAN MECHELEN I,et al.Visualizing distributions of covariance matrices[R].University of Leuwen,Belgium and Columbia University,2011.
    [9]HUANG A,WAND M P.Simple marginally noninformative prior distributions for covariance matrices[J].Bayesian Analysis,2013,8(2):439-452.
    [10]DANIELS M J.Computing posterior distributions for covariance matrices[J].Computing Science and Statistics,1998:192-196.
    [11]BARACHANT A,BONNET S,CONGEDO M,et al.A Brain-Switch using riemannian geometry[C]//Proceedings of the 5th International BCI Conference 2011,Austria:2011:64-67.
    [12]BARACHANT A,BONNET S,CONGEDO M,et al.Common spatial pattern revisited by riemannian geometry[C]//2010IEEE International Workshop on Multimedia Signal Processing(MMSP).Saint-Malo:2010:472-476.
    [13]BARACHANT A,ANDREEV A,CONGEDO M.The riemannian potato:an automatic and adaptive artifact detection method for online experiments using riemannian geometry[C]//TOBI Workshop lV.2013.Sion:2013:19-20.
    [14]FRSTNER W,MOONEN B.A metric for covariance matrices[J].Quo Vadis Geodesia,1999:113-128.
    [15]王翔宇,张引琼.基于MATLAB的时间序列异常检测方法探讨[J].电脑知识与技术,2012,8(4):866-872.
    [16]MABOUDOU-TCHAO E M,AGBOTO V.Monitoring the covariance matrix with fewer observations than variables[J].Computational Statistcs and Data Analysis,2013:99-112.
    [17]MABOUDOU-TCHAO E M,DIAWARA N.A LASSO chart for monitoring the covariance matrix[J].Quality Technology and Quantitative Management,2013,10(1):95-114.
    [18]HAWKINS D M,MABOUDOU-TCHAO E M.Multivariate exponentially weighted moving covariance matrix[J].Technometrics,2008,50(2):155-166.

版权所有:© 2023 中国地质图书馆 中国地质调查局地学文献中心