Banking Observability Monitoring System: An Integrated Framework for Real-Time Financial Transaction Analysis and Anomaly Detection

Author

Smitha T, Niveditha N M

Abstract

 Banking systems process a large number of transactions every day. Monitoring these transactions helps users understand transaction activity, system performance, and changes in transaction patterns.It can be difficult to monitor large amounts of transaction data manually. Unusual transactions, changes in transaction status, and performance issues may not be easily identified without proper monitoring and analysis.The Banking Observability Monitoring System provides an interactive dashboard to analyze transaction data, display important statistics, visualize transaction patterns, and identify unusual transactions using a machine learning technique called Isolation Forest. The system allows users to upload a transaction dataset containing transaction IDs, account IDs, amounts, statuses, and timestamps. The uploaded data is processed and displayed through charts, summaries, and performance indicators. Isolation Forest is used to detect unusual transactions. The system also records logs, measures execution time, and generates a summary report. The system provides a visual summary of transaction activities, displays transaction trends and status distributions, highlights potentially unusual transactions, and helps users monitor application performance. Actual experimental results, such as the number of unusual transactions detected and processing time, should be included if they have been measured.

Keywords

Observability, Isolation Forest, Fraud Detection, Transaction Telemetry, Data Preprocessing, System Metrics.

Full Text:

Download Paper PDF

References

[1]  Kumar, A., & Verma, P. (2023). Real-time transaction monitoring and risk assessment in digital banking. In Proceedings of the IEEE International Conference on FinTech Systems (pp. 112–118).

[2]  Miller, D. K. (2023). Designing Resilient Digital Banking Systems. IEEE Press/Wiley.

[3] Tan, E. S. (2023). Scalable logging and telemetry architectures for high-throughput payment gateways. IEEE Software, 40(2), 64–72.

[4] Ahmed, M. S., & Smith, J. R. (2022). Observability patterns in distributed cloud-native financial services. IEEE Transactions on Service Computing, 15(4), 2100–2112.

[5] Johnston, R. B. (2022). Data preprocessing frameworks for financial anomaly detection. IEEE Access, 10, 45210–45222.

[6] Wang, C. H. (2022). Machine learning approaches to enterprise fraud detection: A survey. IEEE Transactions on Knowledge and Data Engineering, 34(9), 4301–4318.

[7] Zhang, L. et al. (2021). Evaluating microservice observability: Metrics, logs, and traces in practice. In Proceedings of the IEEE International Conference on Software Engineering (pp. 805–816).

[8] Liu, F. T., Ting, K. M., & Zhou, Z.-H. (2008). Isolation forest. In IEEE 8th International Conference on Data Mining (pp. 413–422).

Share your valuable work from Social Media Buttons