An Efficient Way to Detect Credit Card Fraud Using Machine Learning Methodologies

Abstract

Attempted use of combinations of machine learning techniques to detect credit card fraud is presented in this paper. Credit card fraud is increasing day by day every year on a large scale, which results in great loss to organizations, this paper proposes a model to predict whether the credit card presented is fraudulent or not determined by more than 150 attributes per visitor, which have been trained before hand with a dataset. As the data used for the purposes of this paper was highly imbalanced, different sampling techniques have been used to balance the training data. The experiment shows good performance along with accuracy in fraud detection.