Machine Learning for the Exposure of Malicious URLs

In today's era, as per the statistics for world-wide internet users or digital population as on April 2023, there are 5.18 billion internet users worldwide, which goes to 64.6 percent of global population. Of this total, 4.8 billion, or 59.9 percent of the world's population are social media users. [statistica.com/statistics617136/digital-population-worldwide]. Enhancement in technology and internet usage increases the scope for cyber-criminal activities. As the connectivity and the number of users grow, there is a proportional increase in attackers. Vindictive sites or URLs become one of the significant dangers for network protection. Malevolent URLs are most risky danger prompting phishing assault, malware dissemination, Drive-by Downloads, Program Exploits, Data fraud in this computerized world. It's critical to remain cautious and safeguard us on the web. It is vital to be sufficiently mindful and have the option to identify and act against these dangers by not visiting such destinations or giving any private data. In this work we will focus on attacks through malicious URLS, working, preventions and machine learning algorithms used to detect malicious URLs.