USING ADABOOST AND MAJORITY VOTING TO IDENTIFY CREDIT CARD FRAUDULENT ACTIVITY

Abstract

Typically, credit card fraud occurs when a card is taken for an unlawful use or even when the fraudster exploits the credit card details for personal gain. There are a lot of credit card issues in the globe nowadays. The credit card fraud detection system was created in order to identify fraudulent activity. The primary goal of this research is to concentrate on machine learning techniques. The Adaboost algorithm and the random forest method are the algorithms that are employed. The accuracy, precision, recall, and F1-score of the two algorithms are used to determine their outcomes. The confusion matrix is used to plot the ROC curve. After comparing the Random Forest and Adaboost algorithms, the optimum algorithm for detecting fraud is determined by evaluating its accuracy, precision, recall, and F1-score.