采用分段特征表示的异常序列检测算法

doi:10.3778/j.issn.1002-8331.2205-0271

摘要/Abstract

摘要： 时间序列的有监督异常检测方法通常依赖于数据的标签，不仅会消耗大量时间进行数据标注，而且难以适用于无法给定标签的数据集。为解决异常序列检测中的标注问题，提出一种采用分段特征表示的异常序列检测方法。该方法采用分段聚合思想对时间序列进行标准化计算，并得到时序数据的特征表示，可提高无标签时间序列异常检测的可靠性。将表示后的特征划分为异常序列相关特征和无关特征，剪枝异常序列无关特征，可减少这些特征对检测结果的不利影响。为有效量化不同序列之间的差异性，提出一种面向时间权重分析的时间序列相似性度量方法，并构建时间序列的相似度矩阵，用于计算序列之间的相似度，可适用于无标签的时间序列中。在此基础上，根据相似度矩阵来计算每个子序列的异常分数，将其用于异常子序列的判定。通过合成数据集和真实数据集的实验对比表明：该方法节省了计算开销，提高了算法运行的时间效率和异常序列检测的准确率。

关键词: 时间序列, 分段特征表示, 时间权重, 异常序列检测

Abstract: The supervised anomaly detection method of time series usually depends on the label of data, which not only consumes a lot of time for data labeling, but also is difficult to apply to data sets that cannot be given a label. In order to solve the labeling problem in anomaly series detection, an anomaly series detection algorithm based on segmentation feature representation is proposed. This method uses the idea of piecewise aggregation to standardize the calculation of time series, and obtains the characteristic representation of time series data, which can improve the reliability of anomaly detection of unlabeled time series. The expressed features are divided into abnormal series related features and irrelevant features. Pruning abnormal series irrelevant features can reduce the adverse impact of these features on the detection results. In order to effectively quantify the differences between different series, a time series similarity measurement method for time weight analysis is proposed, and the similarity matrix of time series is constructed to calculate the similarity between series, which can be applied to unlabeled time series. On this basis, the anomaly score of each sub-series is calculated according to the similarity matrix, which is used to determine the abnormal sub-series. Finally, the experimental comparison between synthetic data sets and real data sets shows that this method saves the computational overhead, improves the time efficiency of the algorithm and the accuracy of anomaly series detection.The supervised anomaly detection method of time series usually depends on the label of data, which not only consumes a lot of time for data labeling, but also is difficult to apply to data sets that cannot be given a label. In order to solve the labeling problem in anomaly series detection, an anomaly series detection algorithm based on segmentation feature representation is proposed. This method uses the idea of piecewise aggregation to standardize the calculation of time series, and obtains the characteristic representation of time series data, which can improve the reliability of anomaly detection of unlabeled time series. The expressed features are divided into abnormal series related features and irrelevant features. Pruning abnormal series irrelevant features can reduce the adverse impact of these features on the detection results. In order to effectively quantify the differences between different series, a time series similarity measurement method for time weight analysis is proposed, and the similarity matrix of time series is constructed to calculate the similarity between series, which can be applied to unlabeled time series. On this basis, the anomaly score of each sub-series is calculated according to the similarity matrix, which is used to determine the abnormal sub-series. Finally, the experimental comparison between synthetic data sets and real data sets shows that this method saves the computational overhead, improves the time efficiency of the algorithm and the accuracy of anomaly series detection.

Key words: time series, segmentation feature representation, time weight, anomaly series detection

宋春雷, 赵旭俊, 高亚星, 晋广印. 采用分段特征表示的异常序列检测算法[J]. 计算机工程与应用, 2023, 59(9): 262-271.

SONG Chunlei, ZHAO Xujun, GAO Yaxing, JIN Guangyin. Anomaly Series Detection Algorithm Based on Segmentation Feature Representation[J]. Computer Engineering and Applications, 2023, 59(9): 262-271.

参考文献

[1] 滕飞，黄齐川，李天瑞，等.大规模时间序列分析框架的研究与实现[J].计算机学报，2020，43（7）：1279-1292.
TENG F，HUANG Q C，LI T R，et al.An analysis framework for large-scale time series[J].Chinese Journal of Computers，2020，43（7）：1279-1292.
[2] SVN S K，KALIDOS T，KULOTHUNGAN K，et al.Intelligent fuzzy rule-based approach with outlier detection for secured routing in WSN[J].Soft Computing，2020，24（21）：16483-16497.
[3] CAO V L，NICOLAU M，MCDERMOTT J.Learning neural representations for network anomaly detection[J].IEEE Transactions on Cybernetics，2019，49（8）：3074-3087.
[4] 任守纲，张景旭，顾兴健，等.时间序列特征提取方法研究综述[J].小型微型计算机系统，2021，42（2）：271-278.
REN S G，ZHANG J X，GU X J，et al.Overview of feature extraction algorithms for time series[J].Journal of Chinese Computer Systems，2021，42（2）：271-278.
[5] 马超红，翁小清.基于PAA的时间序列早期分类[J].计算机科学，2018，45（2）：291-296.
MA C H，WENG X Q.Early classification of time series based on piecewise aggregate approximation[J].Computer Science，2018，45（2）：291-296.
[6] ESCOBAR S.Anomaly detection principles and algorithms[J].Computing Reviews，2019，60（4）：164-165.
[7] ZENG J，ZHANG L，SHI G，et al.An ARIMA based real-time monitoring and warning algorithm for the anomaly detection[C]//2017 IEEE 23rd International Conference on Parallel and Distributed Systems（ICPADS），2017：469-476.
[8] REN H，YE Z，LI Z.Anomaly detection based on a dynamic Markov model[J].Information Sciences，2017，411（4）：52-65.
[9] GUO C，LI H，PAN D.An improved piecewise aggregate approximation based on statistical features for time series mining[C]//International Conference on Knowledge Science.Berlin，Heidelberg：Springer，2010：234-244.
[10] ZAN C T，YAMANA H.An improved symbolic aggregate approximation distance measure based on its statistical features[C]//Proceedings of the 18th International Conference on Information and Web-based Applications and Services，2016：72-80.
[11] YAHYAOUI H，AL-DAIHANI R.A novel trend based sax reduction technique for time series[J].Expert Systems with Applications，2019，130（9）：113-123.
[12] LE X M T，TRAN T M，NGUYEN H T.An improvement of sax representation for time series by using complexity invariance[J].Intelligent Data Analysis，2020，24（3）：625-641.
[13] THUY H，ANH D T，CHAU V.Efficient segmentation-based methods for anomaly detection in static and streaming time series under dynamic time warping[J].Journal of Intelligent Information Systems，2021，56（3）：121-146.
[14] KIM T Y，CHO S B.Web traffic anomaly detection using C-LSTM neural networks[J].Expert Systems with Applications，2018，106（5）：66-76.
[15] KIEU T，YANG B，JENSEN C S.Outlier detection for multidimensional time series using deep neural networks[C]//2018 19th IEEE International Conference on Mobile Data Management（MDM），2018：125-134.
[16] 胡珉，白雪，徐伟，等.多维时间序列异常检测算法综述[J].计算机应用，2020，40（6）：1553-1564.
HU M，BAI X，XU W，et al.Review of anomaly detection algorithms for multidimensional time series[J].Journal of Computer Applications，2020，40（6）：1553-1564.
[17] ZHOU Y，REN H，LI Z，et al.Anomaly detection via a combination model in time series data[J].Applied Intelligence，2021，51（7）：4874-4887.
[18] REN H，LIAO X，LI Z，et al.Anomaly detection using piecewise aggregate approximation in the amplitude domain[J].Applied Intelligence，2018，48（5）：1097-1110.
[19] 展鹏，陈琳，曹鲁慧，等.核转折点裁剪表示的时间序列异常检测算法[J].计算机工程与应用，2020，56（23）：130-138.
ZHAN P，CHEN L，CAO L H，et al.Time series anomaly detection based on kernel turning points clipped representation[J].Computer engineering and Applications，2020，56（23）：130-138.
[20] LI H.Time works well：dynamic time warping based on time weighting for time series data mining[J].Information Sciences，2021，547（35）：592-608.
[21] AHMED C M，PALLETI V R，MATHUR A P.WADI： a water distribution testbed for research in the design of secure cyber physical systems[C]//Proceedings of the 3rd International Workshop on Cyber-Physical Systems for Smart Water Networks，2017：25-28.
[22] GOH J，ADEPU S，JUNEJO K N，et al.A dataset to support research in the design of secure water treatment systems[C]//The 11th International Conference on Critical Information Infrastructures Security.Cham：Springer，2016：88-99.
[23] CHENG Z，ZOU C，DONG J.Outlier detection using isolation forest and local outlier factor[C]//Proceedings of the Conference on Research in Adaptive and Convergent Systems，2019：161-168.
[24] ALGHUSHAIRY O，ALSINI R，SOULE T，et al.A review of local outlier factor algorithms for outlier detection in big data streams[J].Big Data and Cognitive Computing，2020，5（1）：1-24.
[25] FENG C，TIAN P.Time series anomaly detection for cyber-physical systems via neural system identification and Bayesian filtering[C]//Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining，2021：2858-2867.
[26] AUDIBERT J，MICHIARDI P，FRéDéRIC G，et al.USAD：unsupervised anomaly detection on multivariate time series[C]//The 26th ACM SIGKDD Conference on Knowledge Discovery and Data Mining，2020：3395-3404.