Mutation-Based Multivariate Time-Series Anomaly Generation on Latent Space with an Attention-Based Variational Recurrent Neural Network for Robust Anomaly Detection in an Industrial Control System

被引:1
作者
Jeon, Seungho [1 ]
Koo, Kijong [2 ]
Moon, Daesung [2 ]
Seo, Jung Taek [1 ]
机构
[1] Gachon Univ, Dept Comp Engn Smart Secur, Seongnam Daero 1342, Seongnam Si 13119, South Korea
[2] Elect & Telecommun Res Inst ETRI, Daejeon 34129, South Korea
来源
APPLIED SCIENCES-BASEL | 2024年 / 14卷 / 17期
关键词
anomaly generation; variational Bayes; attention mechanism; recurrent neural network; industrial control system;
D O I
10.3390/app14177714
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
Anomaly detection involves identifying data that deviates from normal patterns. Two primary strategies are used: one-class classification and binary classification. In Industrial Control Systems (ICS), where anomalies can cause significant damage, timely and accurate detection is essential, often requiring analysis of time-series data. One-class classification is commonly used but tends to have a high false alarm rate. To address this, binary classification is explored, which can better differentiate between normal and anomalous data, though it struggles with class imbalance in ICS datasets. This paper proposes a mutation-based technique for generating ICS time-series anomalies. The method maps ICS time-series data into a latent space using a variational recurrent autoencoder, applies mutation operations, and reconstructs the time-series, introducing plausible anomalies that reflect multivariate correlations. Evaluations of ICS datasets show that these synthetic anomalies are visually and statistically credible. Training a binary classifier on data augmented with these anomalies effectively mitigates the class imbalance problem.
引用
收藏
页数:21
相关论文
共 51 条
  • [1] Computer vision and deep learning-based data anomaly detection method for structural health monitoring
    Bao, Yuequan
    Tang, Zhiyi
    Li, Hui
    Zhang, Yufeng
    [J]. STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL, 2019, 18 (02): : 401 - 421
  • [2] Boniol P., 2023, P ADV DAT TECHN EDBT, VVolume 26, DOI [10.48786/edbt.2023.80, DOI 10.48786/EDBT.2023.80]
  • [3] Principal component analysis
    Bro, Rasmus
    Smilde, Age K.
    [J]. ANALYTICAL METHODS, 2014, 6 (09) : 2812 - 2831
  • [4] Supervised Anomaly Detection via Conditional Generative Adversarial Network and Ensemble Active Learning
    Chen, Zhi
    Duan, Jiang
    Kang, Li
    Qiu, Guoping
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2023, 45 (06) : 7781 - 7798
  • [5] Cho K., 2014, P 2014 C EMPIRICAL M, P1724, DOI DOI 10.3115/V1/D14-1179
  • [6] Deep Learning for Anomaly Detection in Time-Series Data: Review, Analysis, and Guidelines
    Choi, Kukjin
    Yi, Jihun
    Park, Changhwa
    Yoon, Sungroh
    [J]. IEEE ACCESS, 2021, 9 : 120043 - 120065
  • [7] Chung J, 2015, ADV NEUR IN, V28
  • [8] SUPPORT-VECTOR NETWORKS
    CORTES, C
    VAPNIK, V
    [J]. MACHINE LEARNING, 1995, 20 (03) : 273 - 297
  • [9] Duchi J, 2011, J MACH LEARN RES, V12, P2121
  • [10] Fabius O., 2015, P 3 INT C LEARN REPR, P1