<![CDATA[The Internet of Things (IoT) integration into banking systems has revolutionized banking operations while also posing threats, including SQL injection (SQLi) attacks. Thus, the defenses of the existing system, such as access control mechanisms, firewalls, and signature-based Intrusion Detection Systems (IDSs), failed to detect both novel and obfuscated SQLi attempts. Hence, this research developed a machine-learning-based detection framework capable of identifying SQLi attacks on IoT-driven banking platforms. The model was trained on a Random Forest (RF) classifier and evaluated in a Python environment. Streamlit was used to deploy the model for real-time prediction, while performance visualization was through the Power BI dashboard. However, the results from the model’s evaluation were highly impressive, with 99.53% accuracy, 99.96% precision, and 98.78% recall. This demonstrated the model's ability to detect both known and unknown SQL patterns. However, the research concluded that combining behavioural analytics with a machine-learning approach is highly effective for securing IoT banking platforms and recommended periodic retraining using a deep-learning approach.]]>
Copyrights © 2026
