A Proactive Method Using Machine Learning Models to Detect Phishing Attacks in Thread Sharing Network
Ajay Kumar K V, R. Deepalakshmi, P S Balasubramaniyamoorthy, K. Murugan · 2024
Phishing attacks present an enduring and pervasive threat in the digital realm, with cybercriminals continuously devising innovative techniques to deceive users and compromise sensitive information. This paper focuses on the crucial task of detecting phishing using advanced machine learning techniques. Recognized as a critical challenge in the digital landscape, phishing detection is imperative due to its potential to undermine trust, compromise security, and inflict financial and reputational damage. The paper employs feature extraction from diverse datasets as a pivotal step in the process, ensuring that relevant attributes are identified and utilized effectively. Moreover, it highlights the system's capability for real-time interpretation of websites, enhancing its responsiveness and efficacy in detecting phishing attempts promptly. The approach integrates machine learning models to enhance accuracy and robustness, achieving a notable accuracy of 97.5%through the utilization of the Gradient Boosting algorithm. This paper contributes to ongoing cybersecurity efforts by offering a multifaceted approach to phishing detection, leveraging the capabilities of machine learning. The findings underscore the importance of continuous research in adapting to evolving phishing tactics and improving the overall security posture of digital environments. Moving forward, the integration of real-time interpretation and adaptive machine learning models promises to further strengthen cybersecurity measures, ensuring a safer digital environment for users worldwide.