You are not logged in. Log in

Add to IET Discover

Immunizing financial organizations against loan and mortgage fraud is a non-trivial problem. It involves the non-trivial identification of valid, novel, potentially useful, and ultimately understandable patterns in very large amounts of data. The critical issue is that we are trying to extract knowledge from data. Depending on the techniques used, the identified knowledge can then be used to: identify fraud in new applications; explain existing fraud; enable logical (as opposed to "graphical") data visualization to aid humans in discovering deeper fraud patterns. An important issue relating to fraud is that as databases grow, fraud identification directly from their contents by humans becomes more and more difficult. A number of approaches have been developed to aid the human in their task. One such approach involves the use of AI techniques such as neural networks and machine induction. In this paper we describe a system based on metaphors taken from the human immune system. This system (referred to as the Artificial Immune System) is a general purpose computer-based learning system which has been applied to applications such as loan and mortgage fraud. (4 pages)