An Ensemble Machine Learning Approach for Enhancing Credit Card Fraud Detection

The rapid advancement of electronic commerce technology has significantly increased credit card usage, making it the most widely used payment method, there are also more fraud cases linked to them. Since fraudulent activities can be concealed in a wide range of acceptable behaviors, it's challenging to identify credit card theft. These days, online credit card payment systems usually create a predictive model to differentiate between transactions that are fraudulent and non-fraudulent. There is a substantial research gap in the development of effective real-time fraud detection systems that are able to spot fraudulent transactions. Here, it is important to design and put into use efficient models and algorithms that can handle large amounts of streaming data and provide efficient, accurate fraud detection. More recently, techniques based on machine learning (ML) have been designed in order to identify credit card frauds; however, due to the imbalance distribution of classes in each dataset, their detection scores still require improvement. To address these challenges, this paper presents An ensemble machine learning approach for enhancing credit card fraud detection. Here, the fraud is found using the AdaBoost and Logistic Regression techniques. The accuracy, true positive rate (TPR), and F1-score of the model are used to evaluate and compare its performance.