Card Fraud Detection Models Using Data Mining Techniques And Patterns

Due to the fast development of e-commerce industry and electronic paymentecosystem, Anti-Fraud systems have a market value. Because of the dissimilarformat of the data (Fraud and Non-Fraud cases), the detection of fraudulenttransactions is difficult to achieve. This paper intends to survey on existing frauddetection models, analyses and compares various popular classifier algorithms thathave been most commonly using in detecting fraud behavior. It focuses on thebenchmark used to assess the classification performance and rank those algorithms.Mostly use Data Mining techniques for credit card fraud detection. The detectiontechniques is mostly based on the methods like Decision Tree, Clusteringtechniques, Neural Networks and Hidden Markov Model, these are evolved indetecting the various credit card fraudulent transactions. This paper presents thesurvey of those techniques and identifies the best fraud cases.

Keywords:

Machine learning , data mining , Fraud , Data mining , Models , Computational efficiency