IJAEMS
IJAEMS
The rapid growth of digital banking generates large volumes of heterogeneous data, creating challenges in financial risk assessment, data quality management, governance, and timely decision-making. This paper proposes an AI-based data engineering approach to integrate banking data and enable intelligent financial risk management. It seamlessly integrates automated data ingestion, data cleansing, quality checks, feature engineering, governance controls, and artificial intelligence-based predictive analysis. Data quality mechanisms detect missing, inconsistent, duplicate, and anomalous records prior to analytical processing, and governance components facilitate secure access to data, traceability of data usage, and controlled access. Machines are used to process the data, detect financial risk patterns, and provide predictive insights for bank decisions using machine learning techniques. A continuous monitoring mechanism assesses the data quality, model performance, and risk changes over time. Experimental results demonstrate 98.67% classification accuracy and a 98.65% weighted F1-score. The framework offers a comprehensive method to integrate data from various sources into a coherent and trustworthy risk intelligence, empowering financial institutions to boost data accuracy, improve risk management governance, maximize risk forecasting power, and facilitate timely and informed riskrating decisions.