Real Time Concept Drift Monitoring System

This paper presented a real-time concept drift monitoring system designed to detect and analyze changes in data streams. The system integrates concept drift detection and anomaly detection with interactive visualization to provide a comprehensive monitoring solution. Unlike traditional approaches that focus only on detection, the proposed system emphasizes usability by presenting concise insights and highlighting key events instead of overwhelming users with raw data. The implementation demonstrates that the system can effectively identify both gradual changes and sudden deviations while maintaining clarity and ease of interpretation. The inclusion of a live monitoring module further enhances its applicability in real-time environments by enabling continuous observation and immediate feedback. Overall, the proposed system improves decision-making by combining detection, visualization, and explainability within a single platform. It provides a practical and user-friendly approach for monitoring dynamic data environments.  

1.               Gama, J., Žliobaitė, I., Bifet, A., Pechenizkiy, M., & Bouchachia, A. (2021). A survey on concept drift adaptation. ACM Computing Surveys.  2.               Lu, J., Liu, A., Dong, F., Gu, F., Gama, J., & Zhang, G. (2020). Learning under concept drift: A review. IEEE Transactions on Knowledge and Data Engineering.  3.               Pang, G., Shen, C., Cao, L., & Hengel, A. (2021). Deep learning for anomaly detection: A review. ACM Computing Surveys.  4.               Sarker, I. H. (2021). Machine learning: Algorithms, real-world applications and research directions. SN Computer Science.  5.               Baena-García, M., del Campo-Ávila, J., Fidalgo, R., Bifet, A., Gavaldà, R., & MoralesBueno, R. (2006). Early drift detection method. Fourth International Workshop on Knowledge Discovery from Data Streams.  6.               Žliobaitė, I. (2010). Learning under concept drift: An overview. arXiv preprint arXiv:1010.4784.