Robust Anomaly Detection Under Contaminated Data: A Comprehensive Evaluation Across PHM Contexts
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
Abstract
Robust anomaly detection under contaminated training data is an important challenge in Prognostics and Health Management (PHM). In semi-supervised anomaly detection, models are typically trained on data assumed to represent normal behavior. In practice, this ``normal'' set often contains an unknown fraction of abnormal or degraded samples, which can harm diagnostic performance. This work presents a comparative evaluation of several techniques designed to mitigate the effects of contaminated training data across four public datasets representative of diverse PHM contexts, spanning tabular and multivariate time-series data, as well as both discrete anomalies and gradual degradation processes. The results show that contamination-mitigating techniques can improve anomaly detection performance over classical baselines when constructing a training set consisting solely of normal instances is not feasible. However, the benefits offered by contamination-mitigating approaches vary according to dataset characteristics. The largest gains are observed on the time-series datasets considered here, suggesting that refinement techniques may offer a clearer advantage over contamination-sensitive baselines in these settings. These gains, however, come at a substantially higher computational cost. The experiments also suggest that the effect of contamination depends not only on its ratio, but also on the structure and distribution of anomalies.
How to Cite
##plugins.themes.bootstrap3.article.details##
Anomaly detection, AD, contamination, contaminated data, autoencoder, AE, refinement, iterative, diagnostic, robust, evaluation
Beggel, L., Pfeiffer, M., & Bischl, B. (2020). Robust anomaly detection in images using adversarial autoencoders. In Machine Learning and Knowledge Discovery in Databases (Vol. 11906, pp. 206–222). Springer. doi: 10.1007/978-3-030-46150-8_13
Candès, E. J., Li, X., Ma, Y., & Wright, J. (2009). Robust principal component analysis? arXiv preprint arXiv:0912.3599. doi: 10.48550/arXiv.0912.3599
Carvalho, T. P., Soares, F. A., Vita, R., Francisco, R. d. P., Basto, J. P., & Alcala, S. G. S. (2022). Artificial intelligence and real-time predictive maintenance in Industry 4.0: A bibliometric analysis. Artificial Intelligence Review, 55, 2127–2170. doi: 10.1007/s10462-021-10082-7
Chandola, V., Banerjee, A., & Kumar, V. (2009, July). Anomaly detection: A survey. ACM Computing Surveys, 41(3), 15:1–15:58. doi: 10.1145/1541880.1541882
Coble, J., & Hines, J. W. (2009). Identifying optimal prognostic parameters from data: A genetic algorithms approach. In Proceedings of the Annual Conference of the Prognostics and Health Management Society (Vol. 1).
Coraddu, A., Oneto, L., Ghio, A., Savio, S., Anguita, D., & Figari, M. (2014, September). Condition based maintenance of naval propulsion plants [dataset]. UCI Machine Learning Repository. doi: 10.24432/C5K31K
del Moral, P., Nowaczyk, S., & Pashami, S. (2022). Filtering misleading repair log labels to improve predictive maintenance models. In Proceedings of the 7th European Conference of the Prognostics and Health Management Society 2022 (Vol. 7, pp. 110–117). Prognostics and Health Management Society. doi: 10.36001/phme.2022.v7i1.3360
Donné, S., Davis, J., Van Utterbeeck, F., & Verbeke, M. (2025, October). Anomaly detection under contaminated data: A weighted iterative refinement framework for health monitoring. In 2025 IEEE 12th International Conference on Data Science and Advanced Analytics. Birmingham, United Kingdom: IEEE. doi: 10.1109/DSAA65442.2025.11248011
Emmott, A. F., Das, S., Dietterich, T., Fern, A., & Wong, W.-K. (2013). Systematic construction of anomaly detection benchmarks from real data. In Proceedings of the ACM SIGKDD Workshop on Outlier Detection and Description (pp. 16–21). New York, NY: Association for Computing Machinery. doi: 10.1145/2500853.2500858
Han, S., Hu, X., Huang, H., Jiang, M., & Zhao, Y. (2022). ADBench: Anomaly detection benchmark. Advances in Neural Information Processing Systems, 35, 32142–32159.
Huang, Y., Tang, Y., VanZwieten, J. H., & Liu, J. (2020). Reliable machine prognostic health management in the presence of missing data. Concurrency and Computation: Practice and Experience, 34(12), e5762. doi: 10.1002/cpe.5762
Kermenov, R., Nabissi, G., Longhi, S., & Bonci, A. (2023). Anomaly detection and concept drift adaptation for dynamic systems: A general method with practical implementation using an industrial collaborative robot. Sensors, 23(6), 3260. doi: 10.3390/s23063260
Kim, M., Yu, J., Kim, J., Oh, T.-H., & Choi, J. K. (2024, October). An iterative method for unsupervised robust anomaly detection under data contamination. IEEE Transactions on Neural Networks and Learning Systems, 35(10), 13327–13339. doi: 10.1109/TNNLS.2023.3267028
Li, Y., Kang, J.-M., & Kim, I.-M. (2025, June). Beyond clean training data: A versatile and model-agnostic framework for out-of-distribution detection with contaminated training data. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 10183–10192). IEEE. doi: 10.1109/CVPR52734.2025.00952
Liu, F. T., Ting, K. M., & Zhou, Z.-H. (2008). Isolation forest. In 2008 Eighth IEEE International Conference on Data Mining (pp. 413–422). IEEE. doi: 10.1109/ICDM.2008.17
Mnih, V., Kavukcuoglu, K., Silver, D., Graves, A., Antonoglou, I., Wierstra, D., & Riedmiller, M. (2013). Playing Atari with deep reinforcement learning. arXiv preprint arXiv:1312.5602. doi: 10.48550/arXiv.1312.5602
Munir, M., Siddiqui, S. A., Dengel, A., & Ahmed, S. (2019). DeepAnT: A deep learning approach for unsupervised anomaly detection in time series. IEEE Access, 7, 1991–2005. doi: 10.1109/ACCESS.2018.2886457
Nectoux, P., Gouriveau, R., Medjaher, K., Ramasso, E., Morello, B., Zerhouni, N., & Varnier, C. (2012, June). PRONOSTIA: An experimental platform for bearings accelerated degradation tests. In Proceedings of the IEEE International Conference on Prognostics and Health Management (pp. 1–8). Denver, CO: IEEE. doi: 10.1109/PHM.2012.6228857
Pang, G., van den Hengel, A., Shen, C., & Cao, L. (2021). Toward deep supervised anomaly detection: Reinforcement learning from partially labeled anomaly data. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (pp. 1298–1308). Association for Computing Machinery. doi: 10.1145/3447548.3467417
Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., ... others. (2019). PyTorch: An imperative style, high-performance deep learning library. In Advances in Neural Information Processing Systems (Vol. 32). Curran Associates.
Qiu, C., Li, A., Kloft, M., Rudolph, M., & Mandt, S. (2022). Latent outlier exposure for anomaly detection with contaminated data. In Proceedings of the 39th International Conference on Machine Learning (Vol. 162, pp. 18153–18167). PMLR.
Scania CV AB. (2016, September). APS failure at Scania trucks [dataset]. UCI Machine Learning Repository. doi: 10.24432/C51S51
Schmidl, S., Wenig, P., & Papenbrock, T. (2022). Anomaly detection in time series: A comprehensive evaluation. Proceedings of the VLDB Endowment, 15(9), 1779–1797. doi: 10.14778/3538598.3538602
Schölkopf, B., Williamson, R. C., Smola, A. J., Shawe-Taylor, J., & Platt, J. C. (2000). Support vector method for novelty detection. In Advances in Neural Information Processing Systems 12 (pp. 582–588).
Shou, H., Lu, G., Pavlovski, M., & Zhou, F. (2025, August). READ: Robust and efficient anomaly detection under data contamination and limited supervision. In Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining (pp. 2586–2596). Toronto, ON, Canada: Association for Computing Machinery. doi: 10.1145/3711896.3737100
Srivastava, N., Mansimov, E., & Salakhudinov, R. (2015). Unsupervised learning of video representations using LSTMs. In Proceedings of the 32nd International Conference on Machine Learning (Vol. 37, pp. 843–852). Lille, France: PMLR.
Su, Y., Liu, R., Zhao, Y., Sun, W., Niu, C., & Pei, D. (2019, August). Robust anomaly detection for multivariate time series through stochastic recurrent neural network. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 2828–2837). Anchorage, AK: Association for Computing Machinery. doi: 10.1145/3292500.3330672
Tan, Y., Tian, H., Jiang, R., Lin, Y., & Zhang, J. (2020, April). A comparative investigation of data-driven approaches based on one-class classifiers for condition monitoring of marine machinery system. Ocean Engineering, 201, 107174. doi: 10.1016/j.oceaneng.2020.107174
Ulmer, M., Zgraggen, J., & Goren Huber, L. (2024, January). A generic machine learning framework for fully unsupervised anomaly detection with contaminated data. International Journal of Prognostics and Health Management, 15(1), 1–12. doi: 10.36001/ijphm.2024.v15i1.3589
Wagner, D., Michels, T., Schulz, F. C. F., Nair, A., Rudolph, M., & Kloft, M. (2023). TimeSeAD: Benchmarking deep multivariate time-series anomaly detection. Transactions on Machine Learning Research.
Welch, P. D. (1967, June). The use of fast Fourier transform for the estimation of power spectra: A method based on time averaging over short, modified periodograms. IEEE Transactions on Audio and Electroacoustics, 15(2), 70–73. doi: 10.1109/TAU.1967.1161901
Wen, J., Ren, J., Zhao, Z., & Chen, X. (2025, August). Robust anomaly detection of rotating machinery with contaminated data. Journal of Dynamics, Monitoring and Diagnostics, 4(3), 170–182. doi: 10.37965/jdmd.2025.855
Yoon, J., Sohn, K., Li, C.-L., Arik, S. O., Lee, C.-Y., & Pfister, T. (2022). Self-supervise, refine, repeat: Improving unsupervised anomaly detection. Transactions on Machine Learning Research.
Yu, J., Kim, M., Kim, J., & Oh, H. (2026). Normality-calibrated autoencoder for unsupervised anomaly detection on data contamination. Neurocomputing, 667, 132249. doi: 10.1016/j.neucom.2025.132249
Zhao, R., Yan, R., Chen, Z., Mao, K., Wang, P., & Gao, R. X. (2019). Deep learning and its applications to machine health monitoring. Mechanical Systems and Signal Processing, 115, 213–237. doi: 10.1016/j.ymssp.2018.05.050
Zhou, C., & Paffenroth, R. C. (2017). Anomaly detection with robust deep autoencoders. In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 665–674). Halifax, NS, Canada: Association for Computing Machinery. doi: 10.1145/3097983.3098052
This work is licensed under a Creative Commons Attribution 3.0 Unported License.
The Prognostic and Health Management Society advocates open-access to scientific data and uses a Creative Commons license for publishing and distributing any papers. A Creative Commons license does not relinquish the author’s copyright; rather it allows them to share some of their rights with any member of the public under certain conditions whilst enjoying full legal protection. By submitting an article to the International Conference of the Prognostics and Health Management Society, the authors agree to be bound by the associated terms and conditions including the following:
As the author, you retain the copyright to your Work. By submitting your Work, you are granting anybody the right to copy, distribute and transmit your Work and to adapt your Work with proper attribution under the terms of the Creative Commons Attribution 3.0 United States license. You assign rights to the Prognostics and Health Management Society to publish and disseminate your Work through electronic and print media if it is accepted for publication. A license note citing the Creative Commons Attribution 3.0 United States License as shown below needs to be placed in the footnote on the first page of the article.
First Author et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 United States License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.