In the more recent years, online purchases have been increased. Therefore, the use of credit cards has also expanded. The issue of security is of great significance in this regard, since credit card fraud has become a challenge in this situation. Currently, several different methods have been proposed to overcome this problem by offering credit card fraud detection strategies. In this research, in order to increase the accuracy of fraud detection, in the feature selection stage, we first calculated the entropy of the mutual information of the features and then provided the obtained values to the genetic algorithm to select the optimal feature vectors. After obtaining the optimal features, the data were fed to the support vector machine classifier for classification. In this step, in order to improve the efficiency of the classifier and better training, the support vector machine parameters, parameter (C) and the Gaussian kernel were adjusted using the particle swarm optimization algorithm. Finally, we were able to achieve 99.6% accuracy using the proposed model.
