Data Science & Machine Learning Methods for Detecting Credit Card Fraud

Abstract

Abstract- Credit Card fraud is the physical loss of a credit card or the loss of sensitive information. For detection there are many machine learning algorithms that can be used. The research demonstrates several algorithms that may be utilized for determining transactions as fraud or genuine. The data set utilized in credit Card Fraud Detection was used in the study. The SMOTE technique was utilized for oversampling because the dataset was extremely imbalanced. Furthermore, feature selection was carried out as well as the set was split into 2 components, test data as well as training data. The algorithms employed in this investigation were Logistic Regression, Naive Bayes, Random Forest, and Multilayer Perceptron. The results demonstrate that each algorithm could be utilized with high accuracy for fraud detection of credit cards. For the detection of additional irregularities, the proposed model could be used.