Abstract

In recent years, the rise of online trading and transactions through the Internet and cloud technologies has brought about increased risks of cyber threats, particularly phishing attacks. These attacks deceive users into revealing sensitive information by impersonating legitimate websites. There various types of prevention of scams here we are going to discuss about the detection of phishing sites Despite efforts such as blacklists and heuristic methods to detect phishing sites, current security technologies often fall short, leading to a growing number of victims.

The anonymous nature of the Internet exacerbates these vulnerabilities. Recognizing the need for more effective defenses, this study proposes a machine learning-based approach using a recurrent neural network to identify phishing URLs. The method was tested with a dataset containing 7900 malicious and 5800 legitimate URLs, demonstrating superior performance compared to existing techniques in detecting malicious URLs. This research aims to enhance protection for users against cyber threats in online environment.