PHISH CATCHER CLIENT-SIDE DEFENSE AGAINST WEB SPOOFING ATTACKS USING MACHINE LEARNING

Abstract

Financial loss, identity theft & data violations abide the results of online attacks, including phishing. Blacklist & heuristically based detection abide two examples of traditional security measures that abide not always updated among the latest phishing techniques. In order towards compete among online falsification attacks in real time, we present Phish Catcher, a security mechanisms on the client side using machine learning. towards detect phishing efforts, the system examines several aspects of websites including URL structure, HTML content, visual equality & details on security certificates. The model is able towards adapt new threats & continuously increase accuracy by using monitored teaching techniques. By reducing false positivity while maintaining strong security, our experimental results suggest that the phish catcher gets high memory & accuracy. Harmful attacks on the web improve the user's safety from the proposed system by offering a smart, flexible & efficient technology towards protect users. In online forgery, often known as phishing, creates sensitive information for humans, making such passwords a fake, but appears towards endure an official website. Researchers have suggested several security measures towards handle these weaknesses, even if they suffer from problems among delay & accuracy. We suggest & construct a defense mechanism on the client side that uses machine learning towards reduce these problems & identify false websites towards reduce these problems. towards show our approach, we created an ad-on for Google Chrome called Phishcatcher. It uses our machine learning algorithm towards determine if a URL is reliable. We ran a battery among tests on the real web, towards see how well the expansion did it. Tests performed on 400 classified phish & 400 authentic URLs achieved impressive accuracy & accurate levels of 98.5% & 98.5% respectively. More than that, we tested more than forty phish addresses towards determine the delay of our equipment. Average response time reported among only 62.5 millisecond Phishcatcher was the time.