A Hybrid Convolutional Neural Network and Support Vector Machine‐Based Credit Card Fraud Detection Model

Abstract

Credit card fraud is a common occurrence in today’s society because the majority of us use credit cards as a form of payment more frequently. This is the outcome of developments in technology and an increase in online transactions, which have given rise to frauds that have caused significant financial losses. In order to detect fraud in credit card transactions, efficient and effective approaches are needed. In this study, we developed a hybrid CNN-SVM model for detecting fraud in credit card transactions. The effectiveness of our suggested hybrid CNN-SVM model for detecting fraud in credit card transactions was tested using real-world public credit card transaction data. The architecture of our hybrid CNN-SVM model was developed by replacing the final output layer of the CNN model with an SVM classifier. The first classifier is a fully connected layer with softmax that is trained using an end-to-end approach, whereas the second classifier is a support vector machine that is piled on top by deleting the final fully connected and softmax layer. According to experimental results, our hybrid CNN-SVM model produced classification performances with accuracy, precision, recall, F1-score, and AUC of 91.08%, 90.50%, 90.34%, 90.41, and 91.05%, respectively.

Real-Time Online Banking Fraud Detection Model by Unsupervised Learning Fusion

“…The accuracy, precision, F1-Score, and recall measurement results of every method are shown in Table 3 In keeping with the outcomes of this study and present state-of-the-art technologies for identifying fraud. For example, our framework obtains an accuracy of 99%, outperforming the CNN-SVM [27], Harris water optimization-RNN [40], and Decision Tree [39] models, producing 90% to 97% accuracy. Moreover, our model has superior precision, recall, and F1-Score ratings of 91%, 97%, and 91%, surpassing the Bidirectional Gated Recurrent Units [41], Decision Tree [38], and K-Nearest Neighbors with CatBoost [39] on these measures.…”

“…The results show that the accuracy of the suggested model is superior and the amount of training error is low. In another study by Berhane et al [27], a hybrid CNN-SVM technique for identifying fraud in credit card transactions was established in this paper, which was evaluated against a real-life available transactional database. Based on the experimental outcomes, the CNN-SVM approach provided classification efficacy with precision, accuracy, and F1-score of 90.50%, 91.08%, and 90.41.…”

Credit Card Fraud Detection Using Improved Deep Learning Models

“…A hybrid CNN-SVM model was developed by [23] by changing the output layer of the CNN model with an SVM classifier. The CNN classifier is fully connected with a softmax layer that is trained using an end-to-end approach, while the second is an SVM classifier that is stacked on top by deleting the final fully connected and softmax layer.…”

“…Secondly, a specialized case of artificial neural networks (ANN) with a distinctive architecture designed to extract progressively complex features of the data at each layer to determine the output. CNN is a feedforward neural network that is built of local connections, shared weights, pooling, and the use of several layers (convolutional and fully connected layers) [23,29,30]. CNN learns the derivative features that are beneficial to the classification results and reduce the interference of human experience with the model [31,32].…”

“…In addition to effectively suppressing noise and focusing on the most discriminative features by leveraging convolutional and pooling operations. So, CNN can improve the model's ability to distinguish between fraudulent and legitimate transactions [23,32,33].…”