Prediction of Credit Card Fraud detection using Extra Tree Classifier and Data Balancing Methods
Vinay Kumar Naramala, Lakshmi Prasanna Ravipudi, Surya Prakesh Reddy Bhavanam, Venkata Nataraj Pokuri, Venkatrama Phani Kumar S, Venkata Krishna Kishore Kolli · 2024
In the digital finance age, increased credit card transactions have led to a rise in fraudulent activities. This study employs machine learning to detect fraud in unbalanced European credit card data. Techniques like SMOTE and Cluster Centroids are utilized for data balancing. Algorithms including Extra Trees, XGBoost, and K-Nearest Neighbors are evaluated using metrics such as accuracy and F1 score. The results show significant improvements, especially with the Extra Trees classifier, emphasizing the method's effectiveness in securing online financial transactions.