An Ensemble Model for Financial Fraud Detection using Deep Learning Techniques to Improve Financial Security

Fraud detection in financial systems is one of the most pressing challenges in today's digital economy. Traditional methods, which often rely on rule-based approaches, struggle to identify advanced fraud patterns. This study proposes an advanced deep learning (DL)-based fraud detection framework that integrates multiple machine learning models to increase prediction accuracy and reduce false positives. The proposed framework combines FinChain-BERT for sequence-based transaction analysis, Multi-Layer Perceptron (MLP) for pattern recognition, Autoencoder for anomaly detection and Deep Neuro-Fuzzy Network for handling uncertain or ambiguous fraud cases. By using these different approaches, a more robust and reliable fraud detection system has been created. Then the effectiveness of the model is demonstrated using a dataset of financial transactions, with evaluation metrics such as precision, accuracy, recall, and F1-score to evaluate the overall performance of the system