The number of merchants and consumers that participate in b2c e-commerce is still growing. Overall fraud rates have stabilized in recent years but for post-payment transactions in the Netherlands the fraud percentage remains unacceptably high. Companies often have a great deal of knowledge about fraudulent orders, and how to recognize them. Fraud prevention is often aided by automated recognition systems that are created through data mining. There have been few studies examining the combination of explicit domain knowledge and data mining. This study analyses the incorporation of domain knowledge in data mining for fraud prediction based on a historical dataset of 5,661 post-payment orders. © Springer-Verlag Berlin Heidelberg 2013.
CITATION STYLE
Polman, T., & Spruit, M. (2013). Integrating Knowledge Engineering and Data Mining in e-commerce Fraud Prediction. In Communications in Computer and Information Science (Vol. 278, pp. 460–466). Springer Verlag. https://doi.org/10.1007/978-3-642-35879-1_56
Mendeley helps you to discover research relevant for your work.