1Department of Computer Engineering, Saveh Branch, Islamic Azad University, Saveh, Iran
2Department of Computer Engineering, Tehran Markaz Branch, Islamic Azad University, Tehran, Iran
Money-Laundering causes a higher prevalence of crime and reduces the desire tending to invest in productive activities. Also, it leads to weaken the integrity of financial markets and decrease government control over economic policy. Banks are able to prevent theft, fraud, money laundering conducted by customers through identification of their clients’ behavioral characteristics. This leads to reduce the banking and credit risks. So there are some systems in order to identify unusual users’ behavior in banking industry that can help different societies. In present study, effective variables are used to determine suspicious behavior in terms of money-laundering from users’ account transactions in an Iranian private bank. Users’ membership degree to clusters is determined using fuzzy clustering method and maximum membership degree is considered as a label for users; also, back propagation neural network is used to identify the model. The results show that the proposed method can detect money-laundering accurately at the bank up to 97%.