Aquila Optimization Algorithm with Random Forest for Real-Time Fraud Detection in Financial Transactions

In recent years, there has been a significant rise in the number of credit card bank transactions as well as the frequency of fraud and card theft. However, as financial crime becomes more challenging task, so improved real-time detection methods is required. This research investigates the use of Machine Learning (ML) approach to identify fraudulent behavior in financial transactions, in real-time applications. Hence, the Aquila Optimization Algorithm with Random Forest (AOA-RF) is recommended in this research as a real-time solution for identifying financial transaction fraud. By using real-time processing, reduction of data, and ML flexibility are required when fraud tendencies arise to properly detect data patterns. By optimizing characteristics and model parameters, the combination of AOA and RF increases real-time financial transaction fraud detection. This hybrid strategy enhances forecast accuracy, decreases false positives, and dynamically adjusts fraud patterns in financial data. According to the results it illustrates that the proposed AOA-RF algorithm have achieved better performances in terms Accuracy of 99.97%, Sensitivity of 99.15% and specificity of 99.54% when compared to existing models such as eXtreme Gradient Boosting (XGBoost), Federated Learning (FL) and Stochastic Gradient Boosting (SGB).