Credit Card Fraudulent Transactions Detection Using Machine Learning

With the rapid growth of the e-commerce industry, the use of credit cards for online purchases has increased significantly. Unfortunately, credit card fraud has also become increasingly prevalent in recent years, creating complications for banks trying to detect fraudulent activity within the credit card system. To overcome this hardship Machine learning plays an eminent role in detecting the credit card fraud in the transactions. Modeling prior credit card transactions with data from ones that turned out to be fraudulent is part of the Card Fraud Detection Problem. In Machine learning the machine is trained at first to predict the output so, to predict the various bank transactions various machine learning algorithms are used. The SMOTE approach was employed to oversample the dataset because it was severely unbalanced. This paper the examines and overview the performance of K-nearest neighbors, Decision Tree, Logistic regression and Random forest, XGBoost for credit card fraud detection. The assignment is implemented in Python and uses five distinct machine learning classification techniques. The performance of the algorithm is evaluated by accuracy score, confusion matrix, f1-score, precision and recall score and auc-roc curve as well.