Detection of Malicious Domains in the Cyberspace using Machine Learning & Deep Learning: A Survey

People have daily access to a wide range of services through the World Wide Web, which serves as a platform for global information exchange. Businesses uses web as a low-cost communication channel for their business promotion and communications. As a result, millions of domains are created and registered every day. This immense development allows cyber criminals to play as well. Cybercriminals use a variety of methods to exploit the web security. Recent cyber security reports states that security breaches are 17% higher in 2021 than in 2020. Cybercriminals use URLs (domain name) as an easy tool to carry out malicious activities. Through email, SMS, and social media sites like Face book and Twitter, malicious URLs are disseminated. As soon as victims click on the URL, fraudsters will steal their personal data or place malware on their computer to get access to other systems. It is very important to detect such malicious domain names in advance using AI based techniques. Researchers have proposed several methods to classify malicious domain names and malicious domain names. Recently they preferred to use machine learning and deep learning techniques for automated classification. This paper presents a recent research work in this field and research issues. This will serve as source of new research in this domain.