Phishing URL Detection: A Real-Case Scenario Through Login URLs

Aim:

         To provide an automated system for the recognition of phishing websites through login URLs

Abstract:

          Phishing attacks continue to pose a significant threat to online security, with attackers employing increasingly sophisticated tactics to trick users into divulging sensitive information. One prevalent method involves the creation of deceptive login URLs that mimic legitimate websites, making it challenging for users to distinguish between genuine and malicious links. This study delves into a real-case scenario of phishing URL detection, focusing on the analysis of login URLs. By examining the characteristics and patterns commonly associated with phishing attempts, we aim to develop effective strategies and tools for the early detection and mitigation of such threats. Through the exploration of emerging technologies and machine learning algorithms, this research seeks to enhance the resilience of individuals and organizations against phishing attacks, ultimately contributing to a safer and more secure online environment aim to improve accuracy.

Existing System:

        The Existing System is existing phishing URL detection methods and finds that the traditional detection methods are difficult to accurately detect phishing URLs. Therefore, we particularly analyzing to detect designed a phishing URL detection method based onthe traditional method. This method given the result is poor performance and accuracy. So we will move the proposed system.

 Problem definition:

              In this traditional method, the accuracy is low so our model will cannot predict the phishing URL accurately.that is a very huge drawback in our existing system

Proposed System:

               Phishing is a social engineering cyberattack where criminals deceive users to obtain their credentials through a login form that submits the data to a malicious server. In this work, we propose machine learning techniques to present a method capable of detecting phishing websites through URL analysis. In most current state-of-the-art solutions dealing with phishing detection, the legitimate class is made up of homepages without including login forms. On the contrary, we use URLs from the login page in both classes because we consider it to be much more representative of a real case scenario and we demonstrate that existing techniques obtain a high false-positive rate when tested with URLs from legitimate login pages. Additionally, we use datasets from different years to show how models decrease their accuracy over time by training a base model with old datasets and testing it with recent URLs

 Advantages:

             Implemented a lexical feature selection from URL to optimize the number of features and the accuracy of their model. removed the less significant ones until they reached an optimal performance.