CREDIT CARD FRAUD DETECTION USING ENSEMBLE LEARNING ALGORITHMS

Abstract

This initiative is centered around the application of intricate algorithms that are driven by ensemble learning to the issue of credit card fraud. The objective is to enhance the reliability and precision of financial system fraud detection. Banks and other financial institutions are extremely concerned about credit card theft as a result of the rapid expansion of online banking and digital payments. The proposed research is distinct from conventional machine learning models in that it investigates the potential of ensemble methods, including XGBoost, Random Forest, Gradient Boosting, and AdaBoost, to identify suspicious activity and more accurately evaluate transaction patterns. Systems that attempt to detect deception are plagued by three significant issues: inaccurate predictions, an abundance of false positives, and an inability to effectively manage datasets that are not well balanced. Ensemble learning is employed to address these challenges. The research indicates that it is crucial to select the appropriate features, prepare the data, and employ evaluation measures such as F1-score, recall, accuracy, and precision in order to detect fraud. Ultimately, our objective is to develop a scalable, intelligent system for detecting deception that can assist companies in reducing expenses and enhancing the security of real-time transfers.