A comprehensive survey of fraud detection methods in credit card based on data mining techniques

Credit cards play a vital role in the financial transactions where it becomes inescapable portion for family, industry and worldwide actions. Although utilizing credit card offers enormous advantages when utilized carefully and liability, fraudulent activities can cause significant credit and financial harm. Several approaches have been suggested to counter the growth of credit card fraudulence, such as k-nearest neighbor, Random Forest, Neural Network, logistic regression, decision tree, hidden Markov model, naïve bayes, Support Vector Machine, or those represent the hybrid approaches etc. The techniques for fraudulence discovery track the user's actions and warn the user when detrimental incident happens. Fraudulence is inhuman behavior and there are an infinite number of various ways of fraud. Thus, based on previous research, various approaches and strategies will be investigated and addressed. In this paper we will learn how to tackle fraud and how to trust the hospital, organizations, and businesses as well insurance companies against fraudulence by Fraud Detection Systems (FDSs) and how to encounter them via Fraud Prevention Systems (FPSs). This survey presents a comprehensive reviewing for the fraudulence matter in credit card, fraud types and challenges that obstruct the performance of FDSs.