E-Mail Fraud Detection - An Organisational Semiotic View

Abstract

Fraud detections methods have become more sophisticated as more cunning ways are being used to commit fraud. Data mining methods have extensively been used in the detection of fraud but the use of ontology based methods is not very prominent and popular. A method based on organisational semiotics which employs semantic analysis and norm analysis is new in this area. Determining the agents, the authorities responsible for decision making, and the affordances, the potential behaviour of these authorities including the rules that govern that behaviour which make up the organisation is important in using these methods in the fraud detection domain. Through this a graphical representation of the relationships between these agents and affordances in the form of an ontology chart can be drawn from which a generic fraud framework can be based.