MUMBAI, India, Feb. 13 -- Intellectual Property India has published a patent application (202641008108 A) filed by Karpagam Academy Of Higher Education; Karpagam Institute Of Technology; Mr. C Sathiskumar; and Surya S, Coimbatore, Tamil Nadu, on Jan. 28, for 'a blockchain-based machine learning framework for android malware detection.'
Inventor(s) include Mr. C Sathiskumar; and Surya S.
The application for the patent was published on Feb. 13, under issue no. 07/2026.
According to the abstract released by the Intellectual Property India: "The increasing prevalence of Android devices and mobile applications has led to a corresponding rise in sophisticated malware attacks that threaten user privacy, financial security, and system integrity. Traditional signature-based malware detection techniques are limited in identifying emerging threats and offer minimal protection against zero-day attacks. To address these challenges, this invention presents AndroMal, an intelligent and secure Android malware detection framework that integrates machine learning with blockchain technology to provide a scalable, reliable, and tamper-proof solution. The proposed system employs the CatBoost algorithm, a high-performance gradient boosting technique, to classify Android applications based on a diverse range of static and behavioral features, including permissions, API calls, intents, and control-flow characteristics extracted using Androguard. This rich feature set enhances the model's capability to detect unknown and obfuscated malware variants with high accuracy. The machine learning model is supported by a decentralized blockchain infrastructure, which securely stores application metadata, detection outputs, and user feedback in an immutable ledger. This prevents adversaries from modifying or forging records, thereby ensuring the integrity and trustworthiness of the detection process. AndroMal further incorporates a real-time monitoring mechanism for continuous scanning of installed and newly downloaded applications. A user-driven feedback loop enhances collaborative threat intelligence, enabling dynamic adaptation to evolving malware patterns. Privacy-preserving measures ensure that sensitive user data is anonymized or protected across all detection and storage stages. The framework is lightweight, scalable, and deployable across mobile devices, enterprise networks, and cloud environments, making it suitable for large-scale security systems. Overall, AndroMal represents a novel and comprehensive approach to Android malware detection, combining the analytical strength of machine learning with the security advantages of blockchain to deliver robust, real-time, and privacy-aware protection against modern mobile threats."
Disclaimer: Curated by HT Syndication.
