Journal of Contemporary Issues in Business and Government

Abstract

Credit card extortion occasions happen often time and then bring about gigantic monetary misfortunes. Culprits can take advantage of some advancement, for example, phishing or Trojan to take another individual's credit card data. Furthermore, strong fraud identification technique is important to recognize the fraud in time when criminals use stolen cards. One strategy is to utilize historical transaction data to obtain normal and fraudulent transactions under the behavioural features of machine learning strategies, and the use of this feature is to verify whether the transaction is valid transaction or invalid transaction. This paper considers four different machine learning algorithms that are Support Vector Machine (SVM), Naïve Bayes, K-Nearest Neighbor (KNN), and Random Forest to discipline the behavior of ordinary transactions and fraud features. Confusion matrix is utilized for estimating the performance analysis of the algorithms. Results obtained from the processing of datasets provide an accuracy of about 99-100%.