Zérosdetect: Phishing URL Detection with Quantum-Driven Zero-shot Learning

Phishing, a prevalent cyber danger in contemporary times, involves the fraudulent impersonation of legitimate websites with the intention of deceiving users into divulging confidential information. The insufficiency of classic phishing detection tools has become apparent as fraudsters continue to develop new strategies; these approaches mostly depend on variables such as URL character sequences, site content, and visual resemblance. The present study highlights the “Zérosdetect” approach, a quantum-powered technique for detecting phishing URLs using zero-shot learning, a robust model that makes use of both the zero-shot learning’s adaptability and the computational advantageous nature of quantum computing, it mitigates the need for extensive previous knowledge of phishing URL attributes. In order to detect previously undiscovered phishing attacks, quantum neural network have been deployed to convert URL data into quantum spaces, therefore using the computational benefits of quantum systems. Quantum layers embedded inside Qnodes are a key part of the present system, calculating gradients at a faster pace, optimizing the network performance by rapidly calculating gradients, making the solution both efficient and forward-thinking. The present study lays the groundwork for future cybersecurity initiatives in the era of quantum computing, enhancing the potential to predict new cases of phishing.