Fraud Detection in Banking Data Using Machine Learning Techniques

Financial fraud in the banking sector has become a growing concern, leading to significant financial losses and security threats. Traditional fraud detection systems rely on rule-based methods, which often fail to detect sophisticated fraudulent activities. This paper proposes a machine learning-based fraud detection system that leverages supervised and unsupervised learning techniques to identify anomalies in banking transactions. The proposed system integrates Random Forest, XGBoost, Support Vector Machine (SVM), and Deep Learning models (LSTM & Autoencoders) to improve fraud detection accuracy. Additionally, Synthetic Minority Over-sampling Technique (SMOTE) and cost-sensitive learning are employed to handle data imbalance issues. Experimental results on real-world banking datasets demonstrate that the proposed approach significantly enhances fraud detection accuracy while reducing false positives.