CREATION OF MACHINE LEARNING-BASED FINANCIAL FRAUD DETECTION SYSTEMS TO ENHANCE THE SECURITY AND RELIABILITY OF DIGITAL FINANCIAL TRANSACTIONS

Objective: This research proposes a novel machine learning-based framework for financial fraud detection that combines ensemble learning techniques with real-time transaction monitoring. Method: Our hybrid approach integrates Random Forest, Gradient Boosting, and Neural Network classifiers to achieve superior detection accuracy while minimizing false positives. Results: Experimental evaluation on real-world datasets demonstrates a fraud detection rate of 97.8% with a false positive rate of only 0.3%, significantly outperforming existing methods. The proposed system offers a scalable solution for enhancing the security and reliability of digital financial transactions. Novelty: The rapid digitization of financial services has created unprecedented opportunities for fraudulent activities, necessitating advanced detection mechanisms.