1. NCCU Academic Hub
  2. Publications
  3. College of Informatics
  4. Theses
  5. Item

Publications-Theses

Article View/Open

Publication Export

Google ScholarTM

NCCU Library

Citation Infomation

Related Publications in TAIR

題名 基於圖神經網路的多變量時間序列相關性異常偵測
Correlation Anomaly Detection in Multivariate Time Series Based on Graph Neural Networks
作者 陳筱詩
Chen, Hsiao-Shih
貢獻者 沈錳坤
Shan, Man-Kwan
陳筱詩
Chen, Hsiao-Shih
關鍵詞 圖神經網路
多變量時間序列
相關性異常偵測
Graph Neural Network
Multivariate Time Series
Correlation Anomaly Detection
日期 2024
上傳時間 4-Sep-2024 15:01:58 (UTC+8)
摘要 隨著感測技術的進步，人們能夠在各應用領域取得時間序列資料。時間序列 資料是隨著時間變化的資訊。這些資料隨著時間推移而不斷變化，並在許多應用 中存在複雜而密切的關係，彼此之間相互影響。在這些變動中，隱藏許多有價值 的訊息，透過捕捉多條時間序列資料中相關性的模式與變化，可預測未來趨勢以 及應對突發事件，得以進行有效的決策支援與風險管理。 本研究旨在進行多變量時間序列相關性的異常偵測。由於多個變量之間存在 複雜的相關性並且隨時間而變化，本論文提出 MCAD-wsGAT (Multivariate Correlation Anomaly Detection - wsGAT) 方法，結合圖神經網路 (Graph Neural Network) 與遞迴神經網路 (Recurrent Neural Networks) 學習與捕捉多變量時間 序列隨時間變化之間的關聯性。根據所學習出的模型預測未來的相關係數，進而 偵測超乎預期的相關性異常 (Correlation Anomaly)。 本研究將實驗應用於伺服器運行中，偵測不同類型異常所造成的相關性異常， 並且觀察與分析時間序列走勢。由實驗結果顯示 MCAD-wsGAT 不僅在相關係數 的預測上有極高的準確率，在相關性異常偵測上也有極佳的效果。
With advancements in sensing technology, time series data—capturing time- related information across various fields—have become increasingly prevalent. These data evolve over time, revealing complex, intertwined relationships that can offer valuable insights. By analyzing patterns and changes in correlations among multiple time series, we can predict future trends and respond to unexpected events, supporting effective decision-making and risk management. We introduce the MCAD-wsGAT (Multivariate Correlation Anomaly Detection - wsGAT) method for anomaly detection in multivariate time series correlations. MCAD- wsGAT combines Graph Neural Networks (GNN) and Recurrent Neural Networks (RNN) to capture temporal correlations and predict future correlation values, enabling the detection of anomalies. We applied this method to server operations to identify correlation anomalies caused by various disruptions. The results show that MCAD-wsGAT effectively models correlation dynamics, reducing MAE loss in predictive models and demonstrating excellent performance in correlation anomaly detection.
參考文獻 [1] P. Esling and C. Agon, Time-Series Data Mining, ACM Computing Surveys (CSUR), vol. 45, no. 1, pp. 1-34, 2012. [2] K. Choi, J. Yi, C. Park, and S. Yoon, Deep Learning for Anomaly Detection in Time-Series Data: Review, analysis, and Guidelines, IEEE Access, vol. 9, pp. 120043-120065, 2021. [3] K. Golmohammadi and O. R. Zaiane, Time Series Contextual Anomaly Detection for Detecting Market Manipulation in Stock Market, in IEEE International Conference on Data Science and Advanced Analytics (DSAA), pp. 1-10, 2015. [4] Y. Lei, F. Jia, J. Lin, S. Xing, and S. X. Ding, An Intelligent Fault Diagnosis Method using Unsupervised Feature Learning towards Mechanical Big Data, IEEE Transactions on Industrial Electronics, vol. 63, no. 5, pp. 3137-3147, 2016. [5] J. Pereira and M. Silveira, Learning Representations from Healthcare Time Series Data for Unsupervised Anomaly Detection, in IEEE International Conference on Big Data and Smart Computing (BigComp), pp. 1-7, 2019. [6] A. Blázquez-García, A. Conde, U. Mori, and J. A. Lozano, A Review on Outlier/Anomaly Detection in Time Series Data, ACM Computing Surveys (CSUR), vol. 54, no. 3, pp. 1-33, 2021. [7] S. Papadimitriou, J. Sun, and C. Faloutsos, Streaming Pattern Discovery in Multiple Time-Series, in Proceedings of the 31st International Conference on Very Large Data Bases (VLDB), pp. 697-708, 2005. [8] E. J. Candès, X. Li, Y. Ma, and J. Wright, Robust Principal Component Analysis?, Journal of the ACM (JACM), vol. 58, no. 3, pp. 1-37, 2011. [9] R. J. Hyndman, E. Wang, and N. Laptev, Large-Scale Unusual Time Series Detection, in IEEE International Conference on Data Mining Workshop (ICDMW), pp. 1616-1619, 2015. [10] M. M. Breunig, H.-P. Kriegel, R. T. Ng, and J. Sander, LOF: Identifying Density- Based Local Outliers, in Proceedings of the ACM SIGMOD International Conference on Management of Data, pp. 93-104, 2000. [11] V. Ishimtsev, A. Bernstein, E. Burnaev, and I. Nazarov, Conformal k-NN Anomaly Detector for Univariate Data Streams, in Conformal and Probabilistic Prediction and Applications, pp. 213-227, 2017. [12] Z. Fu, W. Hu, and T. Tan, Similarity Based Vehicle Trajectory Clustering And Anomaly Detection, in IEEE International Conference on Image Processing, vol. 2, pp. II-602, 2005. [13] A. H. Yaacob, I. K. Tan, S. F. Chien, and H. K. Tan, Arima Based Network Anomaly Detection, in Second International Conference on Communication Software and Networks, pp. 205-209, 2010. [14] Y. Mirsky, T. Doitshman, Y. Elovici, and A. Shabtai, Kitsune: an Ensemble of Autoencoders for Online Network Intrusion Detection, arXiv preprint arXiv:1802.09089, 2018. [15] P. Malhotra, A. Ramakrishnan, G. Anand, L. Vig, P. Agarwal, and G. Shroff, LSTM-based Encoder-Decoder for Multi-Sensor Anomaly Detection, arXiv preprint arXiv:1607.00148, 2016. [16] M. Munir, S. A. Siddiqui, A. Dengel, and S. Ahmed, DeepAnT: A Deep Learning Approach for Unsupervised Anomaly Detection in Time Series, IEEE Access, vol. 7, pp. 1991-2005, 2018. [17] A. Deng and B. Hooi, Graph Neural Network-based Anomaly Detection in Multivariate Time Series, in Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, no. 5, pp. 4027-4035, 2021. [18] H. Zhao et al., Multivariate Time-Series Anomaly Detection via Graph Attention Network, in IEEE International Conference on Data Mining (ICDM), pp. 841-850, 2020. [19] Y. W. Tsao, Anomaly Detection of Correlation Change over Multivariate Time Series, Master’s Thesis, Department of Computer Science, National Chengchi University, 2024. [20] T. N. Kipf and M. Welling, Semi-Supervised Classification with Graph Convolutional Networks, arXiv preprint arXiv:1609.02907, 2016. [21] P. Veličković, G. Cucurull, A. Casanova, A. Romero, P. Lio, and Y. Bengio, Graph Attention Networks, arXiv preprint arXiv:1710.10903, 2017. [22] K. Xu, W. Hu, J. Leskovec, and S. Jegelka, How Powerful Are Graph Neural Networks?, arXiv preprint arXiv:1810.00826, 2018. [23] A. Vaswani et al., Attention Is All You Need, Advances in Neural Information Processing Systems, vol. 30, 2017. [24] M. Grassia and G. Mangioni, wsGAT: Weighted and Signed Graph Attention Networks for Link Prediction, in Complex Networks & Their Applications X: Volume 1, Proceedings of the Tenth International Conference on Complex Networks and Their Applications COMPLEX NETWORKS 2021 10, pp. 369-375, 2022. [25] K. He, X. Zhang, S. Ren, and J. Sun, Deep Residual Learning for Image Recognition, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770-778, 2016. [26] J. Chung, C. Gulcehre, K. Cho, and Y. Bengio, Empirical Evaluation of Gated Recurrent Neural Networks On Sequence Modeling, arXiv preprint arXiv:1412.3555, 2014. [27] Z. Li et al., Multivariate Time Series Anomaly Detection and Interpretation using Hierarchical Inter-Metric and Temporal Embedding, in Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, pp. 3220- 3230, 2021.
描述 碩士
國立政治大學
資訊科學系
111753201
資料來源 http://thesis.lib.nccu.edu.tw/record/#G0111753201
資料類型 thesis
dc.contributor.advisor 沈錳坤zh_TW
dc.contributor.advisor Shan, Man-Kwanen_US
dc.contributor.author (Authors) 陳筱詩zh_TW
dc.contributor.author (Authors) Chen, Hsiao-Shihen_US
dc.creator (作者) 陳筱詩zh_TW
dc.creator (作者) Chen, Hsiao-Shihen_US
dc.date (日期) 2024en_US
dc.date.accessioned 4-Sep-2024 15:01:58 (UTC+8)-
dc.date.available 4-Sep-2024 15:01:58 (UTC+8)-
dc.date.issued (上傳時間) 4-Sep-2024 15:01:58 (UTC+8)-
dc.identifier (Other Identifiers) G0111753201en_US
dc.identifier.uri (URI) https://nccur.lib.nccu.edu.tw/handle/140.119/153390-
dc.description (描述) 碩士zh_TW
dc.description (描述) 國立政治大學zh_TW
dc.description (描述) 資訊科學系zh_TW
dc.description (描述) 111753201zh_TW
dc.description.abstract (摘要) 隨著感測技術的進步，人們能夠在各應用領域取得時間序列資料。時間序列 資料是隨著時間變化的資訊。這些資料隨著時間推移而不斷變化，並在許多應用 中存在複雜而密切的關係，彼此之間相互影響。在這些變動中，隱藏許多有價值 的訊息，透過捕捉多條時間序列資料中相關性的模式與變化，可預測未來趨勢以 及應對突發事件，得以進行有效的決策支援與風險管理。 本研究旨在進行多變量時間序列相關性的異常偵測。由於多個變量之間存在 複雜的相關性並且隨時間而變化，本論文提出 MCAD-wsGAT (Multivariate Correlation Anomaly Detection - wsGAT) 方法，結合圖神經網路 (Graph Neural Network) 與遞迴神經網路 (Recurrent Neural Networks) 學習與捕捉多變量時間 序列隨時間變化之間的關聯性。根據所學習出的模型預測未來的相關係數，進而 偵測超乎預期的相關性異常 (Correlation Anomaly)。 本研究將實驗應用於伺服器運行中，偵測不同類型異常所造成的相關性異常， 並且觀察與分析時間序列走勢。由實驗結果顯示 MCAD-wsGAT 不僅在相關係數 的預測上有極高的準確率，在相關性異常偵測上也有極佳的效果。zh_TW
dc.description.abstract (摘要) With advancements in sensing technology, time series data—capturing time- related information across various fields—have become increasingly prevalent. These data evolve over time, revealing complex, intertwined relationships that can offer valuable insights. By analyzing patterns and changes in correlations among multiple time series, we can predict future trends and respond to unexpected events, supporting effective decision-making and risk management. We introduce the MCAD-wsGAT (Multivariate Correlation Anomaly Detection - wsGAT) method for anomaly detection in multivariate time series correlations. MCAD- wsGAT combines Graph Neural Networks (GNN) and Recurrent Neural Networks (RNN) to capture temporal correlations and predict future correlation values, enabling the detection of anomalies. We applied this method to server operations to identify correlation anomalies caused by various disruptions. The results show that MCAD-wsGAT effectively models correlation dynamics, reducing MAE loss in predictive models and demonstrating excellent performance in correlation anomaly detection.en_US
dc.description.tableofcontents 第一章 緒論 1 1.1 研究背景 1 1.2 研究動機 3 1.3 研究目的 4 第二章 相關研究 5 2.1 多變量時間序列異常偵測 5 2.2 多變量相關性異常偵測 7 第三章 研究方法 8 3.1 研究架構 8 3.2 Correlation Graph Construction 9 3.3 Prediction Model 10 3.4 Anomaly Detection 18 第四章 實驗與結果 20 4.1 實驗設計 20 4.1.1 資料集介紹 20 4.1.2 評估變量 23 4.1.3 Baseline Methods 24 4.1.4 評估所用的Anomaly Threshold 25 4.2 實驗結果 25 4.2.1 Anomaly Detection結果 25 4.2.2 MCAD-wsGAT與Baseline訓練比較 35 4.2.3 異常事件偵測觀察 46 4.2.4 參數調整 50 第五章 結論與未來建議 52 參考文獻 53zh_TW
dc.format.extent 3637840 bytes-
dc.format.mimetype application/pdf-
dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0111753201en_US
dc.subject (關鍵詞) 圖神經網路zh_TW
dc.subject (關鍵詞) 多變量時間序列zh_TW
dc.subject (關鍵詞) 相關性異常偵測zh_TW
dc.subject (關鍵詞) Graph Neural Networken_US
dc.subject (關鍵詞) Multivariate Time Seriesen_US
dc.subject (關鍵詞) Correlation Anomaly Detectionen_US
dc.title (題名) 基於圖神經網路的多變量時間序列相關性異常偵測zh_TW
dc.title (題名) Correlation Anomaly Detection in Multivariate Time Series Based on Graph Neural Networksen_US
dc.type (資料類型) thesisen_US
dc.relation.reference (參考文獻) [1] P. Esling and C. Agon, Time-Series Data Mining, ACM Computing Surveys (CSUR), vol. 45, no. 1, pp. 1-34, 2012. [2] K. Choi, J. Yi, C. Park, and S. Yoon, Deep Learning for Anomaly Detection in Time-Series Data: Review, analysis, and Guidelines, IEEE Access, vol. 9, pp. 120043-120065, 2021. [3] K. Golmohammadi and O. R. Zaiane, Time Series Contextual Anomaly Detection for Detecting Market Manipulation in Stock Market, in IEEE International Conference on Data Science and Advanced Analytics (DSAA), pp. 1-10, 2015. [4] Y. Lei, F. Jia, J. Lin, S. Xing, and S. X. Ding, An Intelligent Fault Diagnosis Method using Unsupervised Feature Learning towards Mechanical Big Data, IEEE Transactions on Industrial Electronics, vol. 63, no. 5, pp. 3137-3147, 2016. [5] J. Pereira and M. Silveira, Learning Representations from Healthcare Time Series Data for Unsupervised Anomaly Detection, in IEEE International Conference on Big Data and Smart Computing (BigComp), pp. 1-7, 2019. [6] A. Blázquez-García, A. Conde, U. Mori, and J. A. Lozano, A Review on Outlier/Anomaly Detection in Time Series Data, ACM Computing Surveys (CSUR), vol. 54, no. 3, pp. 1-33, 2021. [7] S. Papadimitriou, J. Sun, and C. Faloutsos, Streaming Pattern Discovery in Multiple Time-Series, in Proceedings of the 31st International Conference on Very Large Data Bases (VLDB), pp. 697-708, 2005. [8] E. J. Candès, X. Li, Y. Ma, and J. Wright, Robust Principal Component Analysis?, Journal of the ACM (JACM), vol. 58, no. 3, pp. 1-37, 2011. [9] R. J. Hyndman, E. Wang, and N. Laptev, Large-Scale Unusual Time Series Detection, in IEEE International Conference on Data Mining Workshop (ICDMW), pp. 1616-1619, 2015. [10] M. M. Breunig, H.-P. Kriegel, R. T. Ng, and J. Sander, LOF: Identifying Density- Based Local Outliers, in Proceedings of the ACM SIGMOD International Conference on Management of Data, pp. 93-104, 2000. [11] V. Ishimtsev, A. Bernstein, E. Burnaev, and I. Nazarov, Conformal k-NN Anomaly Detector for Univariate Data Streams, in Conformal and Probabilistic Prediction and Applications, pp. 213-227, 2017. [12] Z. Fu, W. Hu, and T. Tan, Similarity Based Vehicle Trajectory Clustering And Anomaly Detection, in IEEE International Conference on Image Processing, vol. 2, pp. II-602, 2005. [13] A. H. Yaacob, I. K. Tan, S. F. Chien, and H. K. Tan, Arima Based Network Anomaly Detection, in Second International Conference on Communication Software and Networks, pp. 205-209, 2010. [14] Y. Mirsky, T. Doitshman, Y. Elovici, and A. Shabtai, Kitsune: an Ensemble of Autoencoders for Online Network Intrusion Detection, arXiv preprint arXiv:1802.09089, 2018. [15] P. Malhotra, A. Ramakrishnan, G. Anand, L. Vig, P. Agarwal, and G. Shroff, LSTM-based Encoder-Decoder for Multi-Sensor Anomaly Detection, arXiv preprint arXiv:1607.00148, 2016. [16] M. Munir, S. A. Siddiqui, A. Dengel, and S. Ahmed, DeepAnT: A Deep Learning Approach for Unsupervised Anomaly Detection in Time Series, IEEE Access, vol. 7, pp. 1991-2005, 2018. [17] A. Deng and B. Hooi, Graph Neural Network-based Anomaly Detection in Multivariate Time Series, in Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, no. 5, pp. 4027-4035, 2021. [18] H. Zhao et al., Multivariate Time-Series Anomaly Detection via Graph Attention Network, in IEEE International Conference on Data Mining (ICDM), pp. 841-850, 2020. [19] Y. W. Tsao, Anomaly Detection of Correlation Change over Multivariate Time Series, Master’s Thesis, Department of Computer Science, National Chengchi University, 2024. [20] T. N. Kipf and M. Welling, Semi-Supervised Classification with Graph Convolutional Networks, arXiv preprint arXiv:1609.02907, 2016. [21] P. Veličković, G. Cucurull, A. Casanova, A. Romero, P. Lio, and Y. Bengio, Graph Attention Networks, arXiv preprint arXiv:1710.10903, 2017. [22] K. Xu, W. Hu, J. Leskovec, and S. Jegelka, How Powerful Are Graph Neural Networks?, arXiv preprint arXiv:1810.00826, 2018. [23] A. Vaswani et al., Attention Is All You Need, Advances in Neural Information Processing Systems, vol. 30, 2017. [24] M. Grassia and G. Mangioni, wsGAT: Weighted and Signed Graph Attention Networks for Link Prediction, in Complex Networks & Their Applications X: Volume 1, Proceedings of the Tenth International Conference on Complex Networks and Their Applications COMPLEX NETWORKS 2021 10, pp. 369-375, 2022. [25] K. He, X. Zhang, S. Ren, and J. Sun, Deep Residual Learning for Image Recognition, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770-778, 2016. [26] J. Chung, C. Gulcehre, K. Cho, and Y. Bengio, Empirical Evaluation of Gated Recurrent Neural Networks On Sequence Modeling, arXiv preprint arXiv:1412.3555, 2014. [27] Z. Li et al., Multivariate Time Series Anomaly Detection and Interpretation using Hierarchical Inter-Metric and Temporal Embedding, in Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, pp. 3220- 3230, 2021.zh_TW