CREDIT CARD FRAUD TRANSACTION DETECTION USING OUTLIER DETECTION MODELS WITH NEURAL NETWORK

Recent developments in e-commerce and telecommunications have increased the use of credit cards in both online and daily transactions. However, credit card fraud is on the rise, causing huge losses for financial institutions every year. Developing effective fraud detection systems is critical to mitigating these losses, but is difficult as most credit card datasets are highly unstable. In addition, credit card fraud using traditional machine learning algorithms is ineffective because its design includes a static map from input vectors to output vectors. As a result, they cannot change the purchasing habits of their credit card customers. This paper presents an effective method for credit card fraud detection using a neural network classifier and a data resampling method. The cluster group was adopted using neural network as a learner based on adaptive learning. The effectiveness of the proposed method is demonstrated using publicly available real-world credit card transaction datasets. The performance of the proposed approach is benchmarked against the following algorithms: Logistic Regression, support vector machine (SVM), multilayer perceptron (MLP), decision tree, traditional XGBoost and MLP.