JJEM: Special Edition 2 (December 2024)
JJEM: Special Edition 2 (December 2024)
2024-12-08
Online Payment Fraud Detection using Machine Learning with XGBoost Classifier.
Dr. Harish B. G, Sukruth K, Sachin Mulagund
Each year, online transaction fraud costs people and financial organizations billions of dollars, making it a serious criminal offense. Highlights the essential role played by financial institutions in identifying and mitigating fraudulent activities; Online transaction fraud can be prevented in a more proactive way using machine learning algorithms with higher precision. It is a piece of cake for someone to commit fraud regarding online transactions. Or in this case, the rise of e-commerce and other online sites have brought a plethora number of options for paying online which has also raised the danger level when it comes to getting frauded. You can easily detect the fraud in online transactions and evaluate it using machine learning methods with an increase in fraudulent activities which are reaching high rates. The focus of this project is the approach to regulate fraud detection using supervised machine learning models by analyzing former transactional data. Transactions are categorized into different groups based on transaction type. Subsequently, individual classifiers are trained and models are evaluated for accuracy. The classifier achieving the highest rating can then be selected as one of the top methods for fraud prediction. Utilizing the Kaggle Synthetic Financial Datasets for Fraud Detection dataset curated by Edgar Lopez-Rojas, we have employed a XGBoost classifier Machine Learning model for detecting fraudulent transactions. An in-depth comparison of these algorithms is conducted to determine the most effective solution.
Payment, Fraud, XGBoost classifier, Transactions