MUMBAI, India, June 30 -- Intellectual Property India has published a patent application (202641073608 A) filed by Vellore Institute Of Technology on June 13, 2026, for “a Password Analysis System For Evaluating Vulnerability Of A Password Beyond Entropy-Based Complexity Assessment”.
Inventors include Dr M Umamaheswari; Saanvi Goel; Vinu Srinivas R; Nithin Aswath Udumalpet Sureshkumar; and Eashwar Shankar.
The application for the patent was published on June 26, 2026, under issue no. 26/2026.
Abstract: The present invention relates to a password analysis system and corresponding method are disclosed for evaluating password vulnerability beyond conventional entropy-based complexity measures. The system includes a password input interface, a structural feature extraction module for deriving features such as length, character class presence, and uniqueness ratio, and a geometric vulnerability modelling module that maps password characters onto a keyboard coordinate space (e.g., QWERTY layout) to compute a keyboard-distance metric indicative of spatial typing patterns, including keyboard-walk vulnerabilities. A semantic reuse detection module normalizes visually obfuscated characters using a substitution mapping and compares the resulting string with a corpus of weak root expressions to identify semantic reuse. A classifier processes a combined feature vector comprising structural features, the keyboard-distance metric, and a semantic reuse indicator to generate a password strength classification along with a confidence score. The system thereby identifies vulnerabilities arising from keyboard adjacency and semantic reuse even in passwords that satisfy traditional complexity requirements, and outputs a predicted strength label and associated confidence level.
Disclaimer: Curated by HT Syndication.
