Time series anomaly detection method based on autoencoder and HMM

doi:10.11772/j.issn.1001-9081.2019091631

Journal of Computer Applications ›› 2020, Vol. 40 ›› Issue (5): 1329-1334.DOI: 10.11772/j.issn.1001-9081.2019091631

• Data science and technology • Previous Articles Next Articles

Time series anomaly detection method based on autoencoder and HMM

HUO Weigang, WANG Huifang

School of Computer Science and Technology, Civil Aviation University of China, Tianjin 300300, China

Received:2019-09-24 Revised:2019-10-19 Online:2020-05-10 Published:2020-05-15
Contact: HUO Weigang, born in 1978, Ph. D., associate professor. His research interests include data mining, fuzzy classification.
About author:HUO Weigang, born in 1978, Ph. D., associate professor. His research interests include data mining, fuzzy classification.WANG Huifang, born in 1993, M. S. candidate. Her research interests include data mining.
Supported by:
This work is partially supported by the Civil Aviation Joint Research Fund of Committee of National Natural Science Foundation of China and Civil Aviation Administration of China (U1633110), the Special Fund for Civil Aviation University of China of Fundamental Research Funds for the Central Universities (3122019190).

基于自编码器和隐马尔可夫模型的时间序列异常检测方法

霍纬纲, 王慧芳

中国民航大学计算机科学与技术学院，天津 300300

通讯作者: 霍纬纲(1978—)
作者简介:霍纬纲(1978—)，男，山西洪洞人，副教授，博士，CCF会员，主要研究方向：数据挖掘、模糊分类；王慧芳(1993—),女，山西大同人，硕士研究生，主要研究方向：数据挖掘。
基金资助:
国家自然科学基金委员会-中国民用航空局民航联合研究基金资助项目（U1633110）；中央高校基本科研业务费项目中国民航大学专项（3122019190）。

Abstract

Abstract:

To solve the issue that the existing symbolic methods of anomaly detection based on Hidden Markov Model (HMM) cannot well represent the original time series, an Autoencoder and HMM-based Anomaly Detection (AHMM-AD) method was proposed. Firstly, the time series samples were segmented by sliding window, and several time series segmented sample sets were formed according to the positions of the segmentations, and the autoencoder of each segmentation was trained by the segmented sample set of different positions on the normal time series. Then, the low-dimensional feature representation of each segmented time series sample was obtained by using the autoencoder, and through K-means clustering of low-dimensional feature representation vector sets, the symbolization of time series sample sets was realized. Finally, the HMM was generated based on the symbol sequence set of the normal time series, and the abnormal detection was carried out according to the output probability values of the test samples on the established HMM. The experimental results on multiple common benchmark datasets show that AHMM-AD improves the accuracy, recall rate, and F1 value by 0.172, 0.477 and 0.313 respectively compared to those of the HMM-based time series anomaly detection model, and has 0.108, 0.450 and 0.319 increasement in these three aspects respectively compared with the autoencoder-based time series anomaly detection model. The experimental results illustrate that AHMM-AD method can extract the nonlinear features in time series, solve the problem that the time series cannot be well represented during the symbolization process of existing HMM-based time series modeling, and also improve the performance of time series anomaly detection.

Key words: autoencoder, symbol sequence, Hidden Markov Model (HMM), anomaly detection, time series

摘要：

针对已有基于隐马尔可夫模型(HMM)的时间序列异常检测模型的符号化方法不能很好地表征原始时间序列的问题，提出了一种基于自编码器和HMM的时间序列异常检测方法(AHMM-AD)。首先,通过滑动窗口对时间序列样本进行分段，按照分段位置形成若干时间序列分段样本集，由正常时间序列上不同位置的分段样本集训练各个分段的自编码器；然后,利用自编码器得到每个分段时间序列样本的低维特征表示，通过对低维特征表示向量集的K-means聚类处理，实现时间序列样本集的符号化；最后,由正常时间序列的符号序列集生成HMM，根据待测样本在已建HMM上的输出概率值进行异常检测。在多个公共基准数据集上的实验结果显示，AHMM-AD比已有的基于HMM的时间序列异常检测模型在精确度、召回率和F1值分别平均提高了0.172、0.477、0.313，比基于autoencoder的时间序列异常检测模型，在这三方面分别平均提高了0.108、0.450、0.319。实验结果表明，AHMM-AD方法能够提取时间序列中的非线性特征，解决已有HMM建模时间序列符号化过程中不能很好表征时间序列的问题，并在时间序列异常检测性能上也有显著提升。

关键词: 自编码器, 符号化序列, 隐马尔可夫模型, 异常检测, 时间序列

CLC Number:

TP391.4

HUO Weigang, WANG Huifang. Time series anomaly detection method based on autoencoder and HMM[J]. Journal of Computer Applications, 2020, 40(5): 1329-1334.

霍纬纲, 王慧芳. 基于自编码器和隐马尔可夫模型的时间序列异常检测方法[J]. 计算机应用, 2020, 40(5): 1329-1334.

References

1 LI L , HANSMAN R J , PALACIOS R , et al . Anomaly detection via a Gaussian mixture model for flight operation and safety monitoring[J]. Transportation Research Part C: Emerging Technologies, 2016, 64: 45-57.
2 CAO Y , LI Y , COLEMAN S , et al . Adaptive hidden Markov model with anomaly states for price manipulation detection[J]. IEEE Transactions on Neural Networks and Learning Systems, 2015, 26(2): 318-330.
3 ZHANG C , CHEN Y , YIN A , et al . Anomaly detection in ECG based on trend symbolic aggregate approximation[J]. Mathematical Biosciences and Engineering, 2019, 16(4): 2154-2167.
4 RABINER L R . A tutorial on hidden Markov models and selected applications in speech recognition[J]. Proceedings of the IEEE, 1989, 77(2): 257-286.
5 孙群丽,刘长良,甄成刚 . 隐马尔科夫模型在滚动轴承故障诊断中的应用[J]. 热能动力工程, 2018, 33(10):95-100. (SUN Q L, LIU C L, ZHEN C G. Application of hidden Markov model in fault diagnosis of rolling bearing[J]. Journal of Engineering for Thermal Energy and Power, 2018, 33(10):95-100.)
6 刘子豪,李凌,叶枫 . 基于SparkR的水文传感器数据的异常检测方法[J]. 计算机应用, 2019, 39(2): 436-440. (LIU Z H, LI L, YE F. Anomaly detection method for hydrologic sensor data based on SparkR[J]. Journal of Computer Applications, 2019, 39(2): 436-440.)
7 SONG C , LIU K , ZHANG X , et al . An obstructive sleep apnea detection approach using a discriminative hidden Markov model from ECG signals[J]. IEEE Transactions on Biomedical Engineering, 2016, 63(7): 1532-1542.
8 KEYGHOBADI H , SEYEDIN A . Abnormality detection in a landing operation using hidden Markov model[J]. Journal of Computer and Robotics, 2017, 10(1): 31-37.
9 杨慧,毛好好,霍纬纲 . 聚类HMM模型在QAR数据分析中的应用研究[J]. 计算机应用与软件, 2018, 35(1): 85-91. (YANG H, MAO H H, HUO W G. The application of clustering HMM model in QAR data analysis[J]. Computer Applications and Software, 2018, 35(1): 85-91.)
10 LI J , PEDRYCZ W , JAMAL I . Multivariate time series anomaly detection: a framework of hidden Markov models[J]. Applied Soft Computing, 2017, 60: 229-240.
11 谭宏强,牛强 . 基于滑动窗口及局部特征的时间序列符号化方法[J]. 计算机应用研究, 2013, 30(3): 796-798. (TAN H Q, NIU Q. Symbolic representation algorithm for time series based on sliding window and local features[J]. Application Research of Computers, 2013, 30(3): 796-798.)
12 JIA Y , XU M , WANG R . Symbolic important point perceptually and hidden Markov model based hydraulic pump fault diagnosis method[J]. Sensors, 2018, 18(12): No.4460.
13 FU T C , HUNG Y K , CHUNG F L . Improvement algorithms of perceptually important point identification for time series data mining[C]// Proceedings of the IEEE 4th International Conference on Soft Computing and Machine Intelligence. Piscataway: IEEE, 2017: 11-15.
14 ZEGEYE W K , DEAN R A , MOAZZAMI F . Multi-layer hidden Markov model based intrusion detection system[J]. Machine Learning and Knowledge Extraction, 2019, 1(1): 265-286.
15 FARSAD N , RAO M , GOLDSMITH A . Deep learning for joint source-channel coding of text[C]// Proceedings of the 2018 IEEE International Conference on Acoustics, Speech and Signal Processing. Piscataway: IEEE, 2018: 2326-2330.
16 廖俊,周中良,寇英信,等 . 一种基于重要点的时间序列分割方法[J]. 计算机工程与应用, 2011, 47(24): 166-170. LIAO J , ZHOU Z L , KOU Y X , et al . Method for time series segment based on important point[J]. Computer Engineering and Applications, 2011, 47(24): 166-170.
17 吴晓婕,胡占义,吴毅红 . 基于Segmental-DTW的无监督行为序列分割[J]. 软件学报, 2008, 19(9):2285-2292. (WU X J, HU Z Y, WU Y H. Unsupervised behavior sequence segmentation based on Segmental-DTW[J]. Journal of Software, 2008, 19(9):2285-2292.)
18 LEKHA S , SUCHETHA M . Real-time non-invasive detection and classification of diabetes using modified convolution neural network[J]. IEEE Journal of Biomedical and Health Informatics, 2018, 22(5): 1630-1636.
19 SAKURADA M , YAIRI T . Anomaly detection using autoencoders with nonlinear dimensionality reduction[C]// Proceedings of the MLSDA 2014 2nd Workshop on Machine Learning for Sensory Data Analysis. New York: ACM, 2014: 4-11.

[1]	HU Tianjie, HU Wenjun, WANG Shitong. Distribution entropy penalized support vector data description [J]. Journal of Computer Applications, 2021, 41(8): 2212-2218.
[2]	ZHANG Yuanjun, ZHANG Xihuang. Dynamic network representation learning model based on graph convolutional network and long short-term memory network [J]. Journal of Computer Applications, 2021, 41(7): 1857-1864.
[3]	LI Yangying, CHEN Zhijun, ZHANG Zihao, YOU Lan. Raw sugar demand forecasting model for sugar manufacturing enterprise based on modified Elman neural network [J]. Journal of Computer Applications, 2021, 41(7): 2113-2120.
[4]	XIE Yu, JIANG Yu, LONG Chaoqi. Extended isolation forest algorithm based on random subspace [J]. Journal of Computer Applications, 2021, 41(6): 1679-1685.
[5]	YAO Jie, CHENG Chunling, HAN Jing, LIU Zheng. Anomaly detection method based on multi-task temporal convolutional network in cloud workflow [J]. Journal of Computer Applications, 2021, 41(6): 1701-1708.
[6]	LI Yanzhi, FAN Yong, GAO Lin. Anomaly detection of oil drilling water flow based on shape flow [J]. Journal of Computer Applications, 2021, 41(6): 1842-1848.
[7]	LI Guorong, YE Jimin, ZHEN Yuanting. Time series clustering based on new robust similarity measure [J]. Journal of Computer Applications, 2021, 41(5): 1343-1347.
[8]	ZHANG Lingzhe, HUANG Xiangdong, QIAO Jialin, GOU Wangminhao, WANG Jianmin. Two-stage file compaction framework by log-structured merge-tree for time series data [J]. Journal of Computer Applications, 2021, 41(3): 618-622.
[9]	CAO Yang, YAN Qiuyan, WU Xin. Ensemble classification algorithm for imbalanced time series [J]. Journal of Computer Applications, 2021, 41(3): 651-656.
[10]	ZHANG Ya, JIN Xin, JIANG Qian, LEE Shin-jye, DONG Yunyun, YAO Shaowen. Deepfake image detection method based on autoencoder [J]. Journal of Computer Applications, 2021, 41(10): 2985-2990.
[11]	HUANG Xiaoxiang, HU Yongmei, WU Dan, REN Lijie. Early identification and prediction of abnormal carotid arteries based on variational autoencoder [J]. Journal of Computer Applications, 2021, 41(10): 3082-3088.
[12]	LI Ziqiang, WANG Zhengyong, CHEN Honggang, LI Linyi, HE Xiaohai. Video abnormal behavior detection based on dual prediction model of appearance and motion features [J]. Journal of Computer Applications, 2021, 41(10): 2997-3003.
[13]	QIAN Bin, ZHENG Kaihong, CHEN Zipeng, XIAO Yong, LI Sen, YE Chunzhuang, MA Qianli. Time series imputation model based on long-short term memory network with residual connection [J]. Journal of Computer Applications, 2021, 41(1): 243-248.
[14]	XIAO Yong, ZHENG Kaihong, ZHENG Zhenjing, QIAN Bin, LI Sen, MA Qianli. Multi-scale skip deep long short-term memory network for short-term multivariate load forecasting [J]. Journal of Computer Applications, 2021, 41(1): 231-236.
[15]	GAO Shile, WANG Ying, LI Hailin, WAN Xiaoji. Prediction method on financial time series data based on matrix profile [J]. Journal of Computer Applications, 2021, 41(1): 199-207.

Time series anomaly detection method based on autoencoder and HMM

基于自编码器和隐马尔可夫模型的时间序列异常检测方法

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics