Evaluation of Support Vector Machine in Fraud Detection

We are living in a world which is rapidly adopting digital payments systems. Credit card and payments companies are experiencing a very rapid growth in their transaction volume. Along with this transformation, there is also a rapid increase in financial fraud that happens in these payment systems. An effective fraud detection system should be able to detect fraudulent transactions with high accuracy and efficiency The automation of several essential service sector elements like electricity bill payment, insurance, mobile recharge and due to the emergence of e- governance projects, almost every individual is forced to depend on online transactions and mobile banking. Online Transaction often face the challenges such as identity theft, skimming, white plastic, site cloning, false merchant sites, merchant collision and triangulation. These challenges are often perpetrated by fraudsters thrill their victims into illicit transaction. Most of the fraud detection mechanisms, even though they provide effective detection. The aim of this study is to evaluate performance of Support Vector Machine in Fraud Detection System in online transaction system to reduce high faise-positive rate and low accuracy. This work used six hundred (600) transactional cardholders simulated dataset with mix of genuine and fraudulent transactions. Three Hundred and Sixty (360) transactions data was used for training while two hundred and forty (240) was used for testing of the fraud detection system... Support Vector Machine (SVM) classification. The performance of the Machine Learning model was evaluated based on accuracy, precision, specificity, sensitivity, and computation time. The study shed light on the robustness and efficiency of different classifiers in handling high- dimensional data, class imbalance, noise, and computational resources.