PHISHING GUARD: MACHINE LEARNING- POWERED WEB SPOOFING DEFENSE

  • Unique Paper ID: 167314
  • Volume: 11
  • Issue: 3
  • PageNo: 799-806
  • Abstract:
  • This project aims to tackle the ongoing threat of phishing attacks by developing Phishing Guard, a client-side defense tool. The main objective is to leverage machine learning as a foundational element for the effective identification of new web spoofing threats. By concentrating on the client side, the project intends to bolster the overall security against phishing attacks. The focus on machine learning highlights the necessity for an adaptable and intelligent defense system. By integrating machine learning into Phishing Guard, the project aims to equip the tool with the capability to outpace the constantly changing tactics used by phishing attackers. This strategy ensures a more efficient and responsive approach to newly emerging web spoofing threats. Recognizing the growing risk posed by phishing, especially with the rise in online activities, this project emphasizes the critical need to combat web spoofing. The development of Phishing Guard is positioned as an essential measure to protect both user privacy and organizational security amidst increasing phishing threats. Unlike traditional server-side solutions that have inherent limitations, Phishing Guard adopts a client-side protection strategy. This choice allows users to gain from a comprehensive defense tool without requiring changes to the targeted websites. This client-side emphasis aims to overcome the limitations of conventional server-side solutions. Phishing Guard is designed with the end-user in mind, particularly those who are frequently targeted by phishing attacks. The tool provides significant advantages by enhancing online safety, drastically reducing the risk of identity theft, and preventing fraud through the effective detection of malicious URLs. By focusing on user protection, Phishing Guard becomes a valuable resource in strengthening individuals against the widespread threat of phishing attacks. We enhanced our anti-phishing tool by integrating Support Vector Machine, XGBoost, and a Stacking Classifier, thereby augmenting the system's capabilities. Additionally, a Flask framework with SQLite was implemented, providing streamlined signup and signin processes for user testing and input validation.

Cite This Article

  • ISSN: 2349-6002
  • Volume: 11
  • Issue: 3
  • PageNo: 799-806

PHISHING GUARD: MACHINE LEARNING- POWERED WEB SPOOFING DEFENSE

Related Articles