Fraud detection involves identifying a fraud as quickly as possible once it has been perpetrated. It requires a tool that is intelligent enough to adapt to criminals' strategies and ever-changing tactics to commit fraud. This paper presents an automated credit card fraud (CCF) detection system based on neural network technology and rule-based component. The Self-Organizing Map (SOM) algorithm was used to create a model of a typical cardholder's behavior and analyze the features of transactions, thus detecting fraudulent transactions. An Artificial Neural Network (ANN) trained with the unsupervised learning method was applied to the data to generate models. An approach was developed to CCF detection that utilizes four clusters (instead of the usual two-stage model normally used in fraud detection algorithms) to reduce the erroneous classification of legitimate transactions as fraudulent and to ensure a more accurate result.

Fraud detection methods are continuously being developed to checkmate criminals who also adopt new strategies regularly. The development of new fraud detection methods is made more difficult due to the severe limitations imposed by restricted information flow about the outcome of fraud detection efforts. Data sets are not made available and results are often not disclosed to the public. The fraud cases have to be detected from the available huge data sets such as the logged data and user behavior. Currently, fraud detection has been implemented by a number of methods such as data mining, statistics, and artificial intelligence. Fraud is discovered from anomalies in data and patterns. The types of frauds include credit card frauds, telecommunication frauds and computer intrusion.

Credit card can be identified as a small plastic card that can be used to buy goods and services and pay for them later. One of the most important and challenging problems for a payment system and its members is the credit card fraud—the illegal use of credit cards by third parties. Credit card fraud is perpetuated in various ways, and it is based on unauthorized write-off of funds from accounts of cardholders. Credit Card Fraud (CCF) can be broadly categorized as application, `missing in post', stolen/lost card, counterfeit card and `cardholder not present' fraud [1].

Science and Technology Journal, Credit Card Fraud Detection, Artificial Neural Networks, ANN, Self-Organizing Map Algorithm, SOM, Data Sets, Counterfeit Card, Fraudulent Transaction, Graphical User Interface, GUI, Llogged Data, Computer Intrusion.