AI-Powered Cybersecurity Risk Scoring for Financial Institutions Using Machine Learning Techniques

Financial institutions are confronted with an expanding range of cybersecurity risks in an increasingly digital financial environment, endangering sensitive client information and business continuity. This paper proposes an Artificial Intelligence (AI)-powered cybersecurity risk scoring model for financial institutions using machine learning (ML) techniques applied to the Lending Club dataset. The approach includes a robust preprocessing pipeline handling missing values, tokenization, normalization, one-hot encoding, and class balancing with SMOTE to enhance data quality and model fairness. Two classification algorithms, Logistic Regression (LR) and Gradient Boosting (GB), are implemented and evaluated using F1-score, recall, accuracy, and precision. The suggested models perform noticeably better than baseline models like BPSOSVMERT and Random Forest (RF). LR obtained a 99.1% F1- score, 99.6% accuracy, 99.7% precision, and 98.6% recall. GB outperformed all with 99.7% accuracy, 99.9% precision, 98.6% recall, and a 99.3% F1-score. These results highlight the effectiveness of the proposed methodology in delivering accurate and reliable cybersecurity risk predictions for enhanced decision-making in financial institutions.