MUMBAI, India, Aug. 17 -- Intellectual Property India has published a patent application (202641095437 A) filed by Vellore Institute Of Technology on August 06, 2026, for System And Method For Anomaly Detection Of Malicious Urls Using Autoencoder-Based Deep Learning.
Inventors include Ramya G; and Krishnendu Pal.
The application for the patent was published on August 14, 2026, under issue no. 33/2026.
Abstract: The present disclosure proposes a system (100) and method for anomaly detection of malicious Uniform Resource Locators (URLs) using autoencoder-based deep learning without requiring labeled malicious data, applicable to web security systems such as network security, web application firewalls, and URL filtering systems. The system (100) comprises a computing device (102), at least one processor (104), a memory (106), and plurality of modules (108) executed through the processor 104 for malicious URL detection through unsupervised anomaly analysis. The system (100) improves the performance of existing URL detection systems, offering a robust solution for high-volume traffic environments with minimal latency. The system (100) maintains flexibility in detecting various types of malicious URLs, ensuring broad applicability to diverse cyber threats such as phishing, malware, and data breaches.
Disclaimer: Curated by HT Syndication.
