Abstract
The Internet of Things (IoT) has transformed financial services through interconnected devices that facilitate digital banking, mobile payments, and automated financial transactions. However, the increasing adoption of IoT technologies has exposed financial infrastructures to cybersecurity threats, including Distributed Denial-of-Service (DDoS), botnet attacks, unauthorized access, and data breaches. Traditional intrusion detection systems often lack the adaptability required to identify evolving and previously unseen threats. This study developed the Financial Transaction Intelligent Threat Detection Model (FTITDM), an intelligent machine learning-based intrusion detection framework for securing IoT-enabled financial transactions. The framework integrates five functional layers comprising IoT devices, network communication, data collection and preprocessing, machine learning detection, and alert and response mechanisms. The study employed three benchmark cybersecurity datasets, namely CICIDS2017, Bot-IoT, and UNSW-NB15, comprising 600,000 records, with 80% allocated for training and 20% for testing. Decision Tree, Random Forest, FTITDM, and Deep Neural Network algorithms were implemented and comparatively evaluated using accuracy, precision, recall, and F1-score. Experimental results revealed that FTITDM achieved the best performance, recording 99.17% accuracy, 98.96% precision, 99.05% recall, and 99.00% F1-score. It also achieved an average detection rate of 99.04% and a false positive rate of 0.89% across the three datasets. These findings demonstrate the effectiveness of FTITDM in detecting diverse cyber threats while maintaining computational efficiency. The study concludes that the proposed framework provides a promising approach to strengthening the security, reliability, and resilience of IoT-enabled financial transaction systems.

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Copyright (c) 2026 Tech-Sphere Journal for Pure and Applied Sciences