MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641076890 A) filed by Sr University on June 22, 2026, for An Enhanced Conditional Generative Adversarial Networks Suicidal Risk Identification System From Social Networking Sites Using Text Similarity Measures.
Inventors include R. Yogesh; and Dr. V. Shobha Rani.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: AN ENHANCED CONDITIONAL GENERATIVE ADVERSARIAL NETWORKS SUICIDAL RISK IDENTIFICATION SYSTEM FROM SOCIAL NETWORKING SITES USING TEXT SIMILARITY MEASURES The present invention discloses an Enhanced Conditional Generative Adversarial Network (CGAN) Suicidal Risk Identification System that analyzes social networking site data using integrated text similarity measures. The system comprises a generator that produces synthetic social media posts conditioned on suicidal risk levels, thereby augmenting scarce high-risk examples and balancing imbalanced datasets. A discriminator enhanced with semantic, lexical, and contextual similarity metrics evaluates the authenticity and risk classification of posts, enabling recognition of subtle and indirect suicidal ideation patterns often missed by conventional models. This dual approach—synthetic data generation combined with advanced text similarity integration—effectively addresses challenges of data scarcity and semantic complexity in noisy social media environments. The invention improves accuracy, sensitivity, and specificity in suicidal risk detection, reduces false positives and negatives, and is adaptable across diverse social networking platforms. The proposed framework provides a robust, context-aware, and scalable solution for real-time suicide risk monitoring, offering significant utility in mental health research, clinical support, and public health interventions.
Disclaimer: Curated by HT Syndication.
