Automated Regulatory (RegTech) Compliance Monitoring in the Financial Landscape Through Machine Learning Models

Regulatory Technology (RegTech) has become an imperative solution in automated compliance monitoring in the financial sector, especially in detecting fraud in imbalanced data of transactions. This paper presents a machine learning-based Regulatory (RegTech) compliance monitoring system based on the Credit Card Fraud Detection (CCFD) dataset. Data integrity and consistency were ensured by conducting comprehensive data preprocessing solutions, such as missing value management, duplicate elimination, and categorical transformation, feature selection, and normalization using StandardScaler. Class imbalance was compensated using Synthetic Minority Over-sampling Technique (SMOTE). The data was split into training and testing subsets. Convolutional Neural Network (CNN) and Random Forest (RF) model have been built and tested on the basis of accuracy, precision, recall, and F 1 -score. The experimental findings prove suggested CNN to be better in comparison to RF, as it only shows 99.8% accuracy and 99.9% precision, recall, and F1-score. The results prove the efficiency of deep learning-based solutions in improving automated financial regulatory compliance monitoring systems.