Big Data Analytics Techniques for Credit Card Fraud Detection: A Review

Due to rapid advancement in internet technology, the use of credit cards has dramatically increased and it leads to increase in number of credit card frauds. The enormous collection of data due to human dependence on computers and automated system is not only helpful for researchers but equally valuable to investigators who intend to carry out forensic analysis of data associated with the variety of criminal cases. The conventional methodologies of performing forensic analysis have changed with the emergence of big data because forensic with big data requires more sophisticated tools along with the deployment of efficient frameworks. This paper presents a survey of techniques used in credit card fraud detection and this work provides a comprehensive review of forensic techniques to detect credit card frauds. Based on analysing the factors such as processing speed, latency, fault tolerance, performance and scalability, an evaluation is made about the techniques and proposed that Apache Spark is performing better for implementation of credit card fraud detection system when compared to other techniques.