Comparison of data mining techniques for Money Laundering Detection System

The work of a police analyst, who inspects money laundering cases, is strongly based on ability to find patterns in a large amounts of financial data, consisting of entries describing bank accounts and money transfers. Out of the need for tools simplifying process of analyzing such data, came the Money Laundering Detection System (MLDS). The mechanisms of money laundering and the system itself are shortly presented in this paper. The main part focuses on implemented algorithms that help finding suspicious patterns in the money flow. Two independent sets of experiments were performed focusing on different algorithms implemented in MLDS, in which their performance was measured, compared and summarized.