Published January 31, 2024 | Version v2
Journal article Open

Process Mining Driven by Deep Learning for Anomaly Detection in Intelligent Automation Systems

Creators

Description

Intelligent automation revolutionizes enterprise operations, software orchestration, and financial systems by integrating AI-driven decision-making, real-time workflow optimization, and large-scale automated execution. However, ensuring system security, operational efficiency, and adaptability in such dynamic environments poses significant challenges. Additionally, heterogeneous automation ecosystems, incorporating cloud-based microservices, robotic process automation (RPA), and distributed AI agents, demand a scalable and adaptive anomaly detection paradigm that can effectively operate across multi-domain environments, particularly in the banking and financial sector, where real-time fraud detection and compliance monitoring are critical. This theoretical concept envisions a deep learning-driven process mining methodology continuously evolving alongside automation workflows, offering a proactive approach to anomaly detection in .NET-based enterprise applications. This paradigm employs multi-layered workflow analysis, anomaly inference through graph neural networks (GNNs), deep feature extraction, and reinforcement learning-driven optimization to deliver a scalable, self-adaptive anomaly detection mechanism.

Additionally, the approach integrates semantic workflow analysis, automated event correlation modeling, and multi-objective optimization to refine anomaly classification granularity and predictive modeling accuracy. By addressing high-dimensional event interdependencies, context-aware deviation analysis, and anomaly reasoning, this model aims to establish a resilient and transparent automation security paradigm that enables real-time workflow intelligence, cross-domain adaptability, and self-improving anomaly mitigation strategies for financial risk assessment, automated loan processing, and fraud analytics in banking environments. Future extensions of this theoretical approach will explore Interpretable Machine Learning (IML), adversarial robustness in deep anomaly detection, and blockchain-based anomaly verification to enhance anomaly interpretability, security, and compliance in enterprise automation ecosystems.

Files

JSAER2024-11-1-317-329.pdf

Files (298.1 kB)

Name Size Download all
md5:2038c0f2222490c3fa875bf5b53fa935
298.1 kB Preview Download

Additional details

References

  • [1]. E. Zuidema-Tempel, R. Effing, and J. Hillegersberg, "Bridging the gap between process mining methodologies and process mining practices," in Lecture Notes in Business Information Processing, vol. 448, pp. 70-86, 2022.
  • [2]. H. M. Marín-Castro and E. Tello-Leal, "Event Log Preprocessing for Process Mining: A Review," Applied Sciences, 2021.
  • [3]. L. Pan and H. Zhu, "An intelligent framework for log anomaly detection based on log template extraction," Journal of Cases on Information Technology, vol. 25, no. 1, pp. 1-23, 2023.
  • [4]. M. Landauer, S. Onder, F. Skopik, and M. Wurzenberger, "Deep learning for anomaly detection in log data: A survey," Machine Learning with Applications, vol. 12, Article no. 100470, 2023.
  • [5]. N. Davis, G. Raina, and K. Jagannathan, "A framework for end-to-end deep learning-based anomaly detection in transportation networks," Transportation Research Interdisciplinary Perspectives, vol. 5, Article no. 100112, 2020.
  • [6]. 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. PP, pp. 1-1, 2021.
  • [7]. Z. Chen, J. Liu, W. Gu, Y. Su, and M. R. Lyu, "Experience Report: Deep Learning-based System Log Analysis for Anomaly Detection," arXiv preprint arXiv:2107.05908, 2021.
  • [8]. S. Liu, L. Deng, H. Xu, and W. Wang, "Logbd: A log anomaly detection method based on pretrained models and domain adaptation," Applied Sciences, vol. 13, no. 13, p. 7739, 2023.
  • [9]. A. Burattin, "Streaming Process Mining," in Process Mining Handbook, W. M. P. van der Aalst and J. Carmona, Eds., Lecture Notes in Business Information Processing, vol. 448, Springer, Cham, 2022.
  • [10]. V.-H. Le and H. Zhang, "Log-based Anomaly Detection with Deep Learning: How Far Are We?," arXiv preprint arXiv:2202.04301, 2022.
  • [11]. T. Shi, Z. Zou, and J. Ai, "Software operation anomalies diagnosis method based on a multiple time windows mixed model," Applied Sciences, vol. 13, no. 20, p. 11349, 2023.
  • [12]. L. Herm, C. Janiesch, H. Reijers, and F. Seubert, "From symbolic RPA to intelligent RPA: challenges for developing and operating intelligent software robots," in Lecture Notes in Business Information Processing, vol. 432, pp. 289-305, 2021.
  • [13]. Y. Zhong and A. Lisitsa, "Can process mining help in anomaly-based intrusion detection?," arXiv preprint arXiv:2206.10379, 2022.
  • [14]. W. Gan, L. Chen, S. Wan, J. Chen, and C. Chen, "Anomaly rule detection in sequence data," arXiv preprint arXiv:2111.15026, 2021.
  • [15]. S. Soheily-Khah and Y. Wu, "Ensemble learning using frequent itemset mining for anomaly detection," Computer Science & Information Technology (CS & IT), vol. 9, no. 9, pp. 321-331, 2019.
  • [16]. L. Feremans, V. Vercruyssen, W. Meert, B. Čule, and B. Goethals, "A framework for pattern mining and anomaly detection in multi-dimensional time series and event logs," in Lecture Notes in Computer Science, vol. 12123, pp. 3-20, 2020.
  • [17]. R. Vaudaine, B. Jeudy, and C. Largeron, "Detection of contextual anomalies in attributed graphs," in Lecture Notes in Computer Science, vol. 12653, pp. 338-349, 2021
  • [18]. J. Li and D. Yang, "Research on financial fraud detection models integrating multiple relational graphs," Systems, vol. 11, no. 11, p. 539, 2023.
  • [19]. J. Ko and M. Comuzzi, "Online anomaly detection using statistical leverage for streaming business process events," arXiv preprint arXiv:2103.00831, 2021.