Machine Learning in Policing Counterfeit Websites

Phishing attacks on websites are a serious cyber-crime whereby mimicking the domain name and appearance of official websites with the aim of stealing confidential information like passwords, credit card information with malicious intent to install malware on the victim's machine or use personalized information to target in multiple types of attacks. When individual victims browse the targeted website, the phishers seize their personal information. Phishing affects assorted fields, such as online business, banking and digital marketing, e-commerce which results in various financial losses and theft of personal and important information. The main purpose of this paper is to present a framework to detect phishing websites using various Machine Learning Algorithms and to focus on several Phishing Attacks along with its Classification Techniques. This detection is done using Uniform Resource Locator (URL). There are many characteristics to distinguish URLs from regular website links, the difference between real and phished website is not visible to the human eye. This paper uses various Machine Learning models and compares their accuracy to profile which model gives best possible analysis. This can benefit website owners to determine best possible mechanism to protect and mitigate phishing attacks.