A genetic algorithm based model for chinese phishing E-commerce websites detection

Abstract

We propose a new Chinese phishing e-commerce websites detection model which integrates the URL features and web features of websites. Some unique features of Chinese e-Commerce websites are included and Sequential Minimal Optimization (SMO) algorithm is applied to identify the phishing ecommerce websites. At the same time, we adopt the genetic algorithm (GA) to optimize the detection model. The evaluation results show that the performance of SMO algorithm is better than the baseline model and GA improves the detection accuracy significantly.