MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202621054339 A) filed by Symbiosis International Deemed University on April 28, 2026, for Machine Learning Based System And Method For Real-Time Phishing Website Detection Using Url Feature Analysis.
Inventors include Omkar Mendhe; Himanshu Mahakalkar; Himanshu Ghatole; and Dr. Priya Dasarwar.
The application for the patent was published on June 19, 2026, under issue no. 25/2026.
Abstract: ABSTRACT MACHINE LEARNING BASED SYSTEM AND METHOD FOR REAL-TIME PHISHING WEBSITE DETECTION USING URL FEATURE ANALYSIS A machine learning based system and method for real-time detection of phishing websites using URL feature analysis. The system comprises a data collection module (110) acquiring labeled URL datasets, a preprocessing module (120) performing data cleaning and normalization, a feature extraction module (130) extracting URL-based features including URL length, subdomain count, special characters, domain length, and digit count. A classification module (140) implements Decision Tree Classifier (142) and Random Forest Classifier (144) for phishing detection. The Random Forest Classifier (144) employs ensemble learning through aggregation of multiple decision trees achieving 88% detection accuracy. The system enables real-time phishing detection without dependency on external blacklists, providing effective protection against zero-day phishing attacks through learned pattern recognition. A browser extension interface (172) intercepts URL requests and provides user warnings for potentially malicious websites. [
Disclaimer: Curated by HT Syndication.
