Malicious Domain Detection Based on Decision Tree

Different types of malicious attacks have been increasing simultaneously and become a serious issue for cybersecurity. Most leverage domain URLs as an attack communications medium compromise users into victim phishing or spam. We take advantage machine learning methods to detect the maliciousness automatically using three features: DNS-based, lexical, semantic features. The proposed approach exhibits high performance even with small training dataset. experimental results demonstrate that scheme achieves approximate accuracy 0.927 when random forest classifier.