MUMBAI, India, Jan. 9 -- Intellectual Property India has published a patent application (202521128937 A) filed by Symbiosis International, Pune, Maharashtra, on Dec. 18, 2025, for 'system and method for ransomware detection and recovery using reinforcement learning.'
Inventor(s) include Prof. Firdous Sadaf; Parth Denge; Samir Shendre; and Krishna Khote.
The application for the patent was published on Jan. 9, under issue no. 02/2026.
According to the abstract released by the Intellectual Property India: "A computer-implemented system for ransomware detection and recovery using deep reinforcement learning is disclosed. The system comprises a processor coupled to a memory storing executable instructions. The system includes a behavioral analysis module configured to monitor file system activity data comprising file access frequency, file modification patterns, directory traversal speed, and file entropy changes. A reinforcement learning module analyzes the activity data using a Proximal Policy Optimization algorithm to generate behavioral state vectors representing encryption patterns and anomaly scores. A detection module evaluates the behavioral state vectors against trained policy models to classify activities as normal operations or malicious encryption patterns. A recovery module executes automated file restoration using snapshot-based rollback mechanisms when ransomware is detected. The system achieves detection accuracy exceeding 94 percent with response latency less than one second and completes file recovery in approximately three to fifteen seconds, providing adaptive ransomware defense without requiring predefined signature databases."
Disclaimer: Curated by HT Syndication.
