Year: 2026
Type: article
Abstract: As digital banking, online payments, and e-commerce rapidly evolve, financial transactions are increasingly vulnerable to fraud. Traditional fraud detection systems often produce high false positives and fail to adapt to changing tactics. Our solution employs advanced supervised learning algorithms like Random Forest, XGBoost, and Logistic Regression to accurately analyse transactional data and quickly identify fraudulent activity. By integrating sophisticated data preprocessing, feature engineering, and techniques to manage class imbalances, we significantly improve prediction accuracy. Our results show that these ensemble-based models outperform conventional classifiers, delivering enhanced fraud detection with fewer false alarms, ultimately strengthening transaction security and reducing financial losses in modern digital banking.
Institution: CMR University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W7140138882
FWCI:
Cites:
Cited by:
Related to: