MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641076750 A) filed by Sns College Of Technology on June 21, 2026, for Sentrysoc – Intent-Centric Cognitive Pdf Malware Detection System.
Inventors include Dr. Husna Khouser G; Dharshan R; Jeeva M P; Madhankumar A; Arish S; Dr. M. Siva Ramkumar; and Mrs. B. Sajitha.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: The present invention relates to an advanced cybersecurity system for detecting 5 malicious PDF documents through intent-centric cognitive analysis. The system is 6 designed to identify both known and zero-day malware by analyzing the semantic 7 behavior and execution intent of PDF files rather than relying solely on signature-based 8 or structural detection techniques. It introduces a novel framework that converts low- 9 level PDF components into high-level behavioral representations, enabling deeper 10 understanding of malicious actions. 11 12 The platform incorporates an action extraction module that parses PDF structures, 13 embedded scripts, and event triggers to identify executable elements. These elements are 14 processed by an intent generation engine, which transforms them into semantic 15 descriptions representing operational goals such as hidden execution, data exfiltration, or 16 external payload retrieval. An intent graph model is constructed to represent the sequence 17 and dependencies of actions, enabling graph-based behavioral analysis. 18 19 The system further integrates a neuro-symbolic reasoning engine, combining machine 20 learning models with rule-based inference to enhance detection accuracy and 21 interpretability. It supports anomaly-based zero-day detection by identifying deviations 22 from normal behavioral patterns, ensuring resilience against obfuscated and previously 23 unseen threats. Additionally, an explainable AI module generates human-readable 24 insights, highlighting suspicious components and reasoning behind classification 25 decisions. 26 27 SENTRYSOC provides a scalable and unified platform for real-time threat analysis, 28 supporting deployment in security operations centers, enterprise environments, and 29 research systems. It includes structured data storage for experiment tracking, 30 continuous learning, and threat intelligence integration. By shifting the focus from pattern 31 recognition to intent-driven analysis, the invention significantly improves robustness, 32 adaptability, and transparency in PDF malware detection.
Disclaimer: Curated by HT Syndication.
