Abstract:
The Malicious URL, a.k.a. Malicious Website, is a common and serious threat to cybersecurity. Malicious/Phishing URLs host unsolicited content and are used to perpetrate cybercrimes. Malicious/Phishing URLs may lead to illegitimate access to the user data by adversary.It is imperative to detect them in a timely manner. Traditionally, this is done through the usage of blacklists, Heuristic Classification etc., which cannot be exhaustive, and cannot detect newly generated Malicious/Phishing URLs. Furthermore, these approaches also fall short in detecting the modern URLs such as short URLs, dark web URLs. Even though we are currently interested in just machine learning techniques, but out of all (such as CNN, Random Forest etc.), K-Nearest Neighbors(KNN) and Support Vector Machine (SVM) provided the better results this is because of the effective learning rate and quite suitable for the feature extraction. The proposed classification model is built on sophisticated machine learning methods that not only takes care about the syntactical nature of the URL, but also the semantic and lexical meaning of these dynamically changing URLs. The proposed approach is expected to outperform the existing techniques.

Keywords: Malicious URLs, Black-Listing, Machine Learning, URL Features, Cyber Crime, KNN, SVM.