Detecting and repairing anomalies in business process event logs is essential for maintaining process integrity. Building on the observation that real-world anomalies often exhibit specific patterns and extending our earlier semi-supervised, rule-based approach to detect and repair common basic patterns, this paper presents an anomaly detection and repair framework powered by reinforcement learning. The proposed approach automatically learns an intelligent policy to identify and correct typical basic anomaly patterns, moving beyond the limitations of heuristic, rule-based methods. Furthermore, thanks to the flexibility provided by reinforcement learning, the approach can handle more complex patterns formed by combinations of the basic ones. Extensive evaluation using both synthetic and realworld logs demonstrates that the proposed method outperforms traditional trace alignment, edit distance-based techniques, and unsupervised deep learning in the accuracy of anomalous trace repair. It also shows consistent performance across various anomaly types and offers a pattern categorization capability that baseline methods lack.
- A Reinforcement Learning Framework for Event Log Anomaly Detection and Repair
- J Ko - Queensland University of TechnologyMT Wynn - Queensland University of TechnologyM Comuzzi - Ulsan National Institute of Science and TechnologyFabrizio Maria Maggi - Free University of Bozen-Bolzano
- 2025 7th International Conference on Process Mining (ICPM), pp.1-8
- 9798331585969
- International Conference on Process Mining (Montevideo, 20/10/2025–24/10/2025)
- Institute of Electrical and Electronics Engineers Inc.
- 8
- 9798331585969
(UNIBZ)101958096
991007447180501241 - WOS:001851541200022
- 2-s2.0-105024934291
- Faculty of Engineering
- English
- Conference proceeding
- Ko J, Wynn MT, Comuzzi M, Maggi FM