MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202641041742 A) filed by J Manoj Prabhakar; and Dhaanish Ahmed Institute Of Technology, Coimbatore, Tamil Nadu, on April 1, for 'system and method for real-time multilingual sms spam detection on mobile devices using indicbert.'
Inventor(s) include J Manoj Prabhakar; B Primila; S Saleeha; S Santhoshkumar; Sekar S; Harishwar T; Ragavan T; and V J Amrutha.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "The present invention discloses a system and method for real-time SMS spam detection on mobile devices using a transformer-based multilingual language model. Incoming SMS messages are intercepted, preprocessed, and encoded using a fine-tuned IndicBERT model trained on multilingual datasets. The generated contextual embeddings are processed by a classification layer to determine whether the message is spam or legitimate. The trained model is optimized and deployed using TensorFlow Lite within an Android application, enabling real-time on-device inference while preserving user privacy. The system supports multiple Indian languages and provides improved spam detection accuracy compared to conventional machine learning approaches."
Disclaimer: Curated by HT Syndication.
