Development and Implementation of a system for identifying relationships of suspicious customers in money laundering using social networking functions

Money laundering is a growing issue in the current world trade as it funds illegal activities that can demoralize humanity and civilization. Many research works have been conducted using pre-defined rules and statistical approaches to detect abnormal transactions activities based on customer transactions information. However, current approaches do not consider relationships and association of related customers and/or events. It is believed that the social networking functions can be used to find patterns, associations, or relationships of suspicious customers such as family relations, business relations, common owners, and acquaintance relations that could be associated and involved in illegal money transaction activities.In this project, we will explore various mechanisms and techniques on how to utilize such information in order to extract hidden relationships of suspicious customers. First, we will examine the feasibility of social networking functions to identify hidden relationships and associations on real banking data. Next, we will design and develop rules that can be used to extract hidden relationships of suspicious customers. We will then determine the steps and procedures required for analysing the associations among the customers and transactions. Further, we aim to display those associations and relationships visually to aid human analysis by using the appropriate Graph tools. To further improve our results, our project will employ machine learning algorithms to extract the hidden associations and relationships. Finally, we will evaluate the performance of our techniques using various performance metrics.