MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641079967 A) filed by Sona College Of Technology on June 29, 2026, for A Hybrid Machine Learning-Based System And Method For Real-Time Detection Of Phishing Urls.
Inventors include Karthick K; Selvanathan N; Murali M; Maha Lakshmi S; Kowisha P; and Madhumita V.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: ABSTRACT The present invention relates to a computer-implemented system and method for real-time detection of phishing Uniform Resource Locators (URLs) using a hybrid machine learning framework. The system comprises a URL acquisition module, a preprocessing module, a feature extraction module, a hybrid detection engine, a feature fusion module, a classification and confidence evaluation module, a monitoring and alert generation module, and a data storage and reporting module. The feature extraction module derives lexical, structural, contextual, statistical, domain-related, and security-related attributes from received URLs. The hybrid detection engine integrates a Convolutional Neural Network (CNN), a Transformer-based sequence analysis model, and a Gradient Boosting classifier to generate phishing prediction outputs. The outputs are fused to determine a phishing likelihood score and classify URLs as safe, suspicious, or phishing. The system further generates confidence scores, automated alerts, monitoring information, and threat intelligence reports, thereby enabling real-time detection and mitigation of phishing threats, including previously unseen and dynamically generated malicious URLs.
Disclaimer: Curated by HT Syndication.
