MUMBAI, India, Aug. 17 -- Intellectual Property India has published a patent application (202641065095 A) filed by Srm Trp Engineering College on May 23, 2026, for Deep Trace - Illicit Drug Detection On Social Media Using Distilbert, Efficientnet, And Regex-Based Filtering.
Inventors include Abarna S; Adithya R M; Jagadeesh Kumar P; Kalpanadevi S; and Dr P. Sudhakaran.
The application for the patent was published on August 14, 2026, under issue no. 33/2026.
Abstract: The present invention proposes an artificial intelligence based system called DeepTrace for detecting illicit drug promotion on social media platforms. The system analyzes both textual and visual content shared in social media posts using a multimodal approach. Text data is processed using natural language processing techniques, including transformer-based models and regexbased pattern detection, to identify suspicious language and drug-related slang. Images are analyzed using convolutional neural network models to detect visual patterns associated with drug-related materials. The outputs from the text and image analysis modules are combined through a multimodal decision fusion mechanism to generate a final classification of the content as drug-related or normal. The system provides an automated solution for monitoring large volumes of social media data and helps improve the detection of illicit drug promotion activities.
Disclaimer: Curated by HT Syndication.
