Design Framework for Online Payment Transaction Fraud Detection

Online payment transactions pose a risk of fraudulent transactions. In addition to security measures, fraudsters perform unauthorized online payment transactions. Because, at regular intervals, fraudsters change their fraudulent transaction strategies and come up with more advanced fraudulent techniques, especially attacks on the user’s end. This chapter presents the two proposed frameworks and one proposed model for identifying and mitigating fraudulent online payment transactions. The first framework includes the multifactor authentication technique. The multifactor authentication framework aims to restrict fraudsters from executing fraudulent online payment transactions under any circumstances. The second framework comprises the fraud detection system that relies on the decision support system. The goal of the decision support system framework is to identify anomalies in the current transaction by comparative analysis with the previous transaction pattern of an authorized user. The third model incorporates the deep autoencoder based on a data-driven approach. The deep autoencoder model aims to grab the anomalies in the current transaction by comparing it to the previous transaction pattern of an authorized user. This chapter illustrates the basic foundation of data collection in relation to the proposed two frameworks and one model for identifying and mitigating fraudulent online payment transactions.