With the rapid growth of the population, people have more access to the internet every year. According to this estimate, 66.2% of the global population uses the internet. More precisely, over the decade 2000-2022, internet usage increased by 1,355%. Accordingly, this huge usage provided hackers with the opportunity to carry out malicious activities such as Phishing attack. Phishing is known as social engineering technique through using a fraudulent message to trick a victim into disclosing sensitive information or to install malware on the victim's computer, as in the case of ransomware. A number of methods have been proposed to detect phishing websites. Phishers have evolved their methods to escape detection methods and ML has proven to be one of the most effective methods for detecting these malicious activities. This Thesis discusses the results of ML techniques already widely used by researchers and it will implement an unsupervised technique that can detect phishing attacks efficiently.