Exploring Support Vector Machines and Random Forests to Detect Advanced Fee Fraud Activities on Internet

Abiodun Modupe, Oludayo O. Olugbara, Sunday Olusegun Ojo

Abstract: In this study, we experiment with Support Vector Machines (SVM) and Random Forests (RF), which are two of the state-of-the-art machine learning algorithms. The purpose was to examine their suitability for detecting Advanced Fee Fraud (AFF) activities on internet, which due to its inherent vulnerability is often abused for various criminal activities. A set of cluster features was discovered using global CM algorithm to characterize AFF activities on internet. These features were used to train SVM and RF to classify an e-mail document as either containing AFF related information or not. The results of experiments performed show that both SVM and RF have a satisfactory performance in detecting AFF activities. However, SVM comparatively shows superior result than RF and can effectively detect the dynamic nature of AFF activities on internet.