Cyber Fraud Prediction with Supervised Machine Learning Techniques
Zhoulin Li, Hao Zhang, Mohammad Masum, Hossain Shahriar, Hisham M. Haddad · 2020
With the spread of electronic payments, the number of credit card transactions continues to grow, taking an increasing share of the U.S. payment system. However, this has led to an increasing rate of stolen accounts and losses for Banks. As a result, improved fraud detection has been the goal of reducing Banks' losses. In recent years, with the rapid development of information technology and the progress of machine learning methods, researchers are committed to applying some machine learning methods to financial field. In this paper, we detect credit card fraud transaction from publicly available datasets using Naive Bayes, logical regression, and artificial neural networks on cybersecurity classification problems. The results from our experiments show that for datasets with more balanced for classification, the accuracy of logistic regression was higher than Naive Bayes and neural network algorithms. The Naive Bayes algorithm performed better than logistic regression and neural networks in the dataset with highly imbalanced distribution.