Detecting Credit Card Fraud using Machine Learning

Abstract

Credit card is getting increasingly more famous in budgetary exchanges, simultaneously frauds are likewise expanding. In the past, fraud practitioners were identified using rule-based master frameworks, which ignored a variety of variables, including the outlandishly imbalanced nature of positive and negative cases. Using named information, we provide an approach to fraud detection that uses Convolutional Neural Networks (CNNs) and is based on CNNs. An element lattice speaks to a plethora of interchange information and uses a convolutional neural organization to recognize a large number of idle examples for each of those examples. A considerable business bank's boss presentation is compared with several best-in-class techniques in trials on truly monstrous exchanges. Our objective is to combine CNN with LSTM and Auto-encoder to increase credit card fraud detection while improving the previous models' performance. By using these four models; CNN, AE, LSTM, and AE&LSTM. each of these models is trained by different parameter values highest accuracy has been achieved where the AE model has accuracy = 0.99, the CNN model has accuracy = 0.85, the accuracy of the LSTM model is 0.85, and finally, the AE&L-STM model obtained an accuracy of 0.32 by 400 epoch. It is concluded that the AE classifies the best result between these models.