Malicious URL Detection to Avoid Web Crime Using Machine Learning Techniques

Abstract

Today the foremost necessary concern within the field of cyber security is finding the intense issues that create loss in secure data. In recent years, most offensive strategies are applied by spreading malicious and phishing URLs. An accidental visit to a malicious website will trigger pre-designed criminal activity. The phishing website has evolved as a serious cyber security threat in recent times. Phishing may be a type of online fraud wherever a spoofed website tries to gain access to user's sensitive data by tricking the user into believing that it's a benign website. ML algorithms are one of the effective techniques for malicious website detection. The proposed system detects whether the link is legitimate or malicious not based on varied discriminative options and attributes of the address. This application is enforced with the assistance of Gradient Boosting classifier. The model can find whether the address is safe or unsafe.