Design of an artificial immune system as a novel anomaly detector for combating financial fraud in the retail sector

The retail sector often does not possess sufficient knowledge about potential or actual frauds. This requires the retail sector to employ an anomaly detection approach to fraud detection. To detect anomalies in retail transactions, the fraud detection system introduced in this work implements various salient features of the human immune system. This novel artificial immune system, called CIFD (Computer Immune system for Fraud Detection), adopts both negative selection and positive selection to generate artificial immune cells. CIFD also employs an analogy of the self-Major Histocompatability Complex (MHC) molecules when antigen data is presented to the system. These novel mechanisms are expected to improve the scalability of CIFD, which is designed to process gigabytes or more of transaction data per day. In addition, CIFD incorporates other prominent features of the HIS such as clonal selection and memory cells, which allow CIFD to behave adaptively as transaction patterns change.