MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641080411 A) filed by Cmr Institute Of Technology on June 30, 2026, for System And Method For Automated Detection Of Suicidal Ideation In Social Media Content Using Machine Learning And Natural Language Processing.

Inventors include Sumeet Medi; Dr K. Niranjan Reddy; Katakam Srinivasa Rao; Sadhu Malli Babu; V Surekha; and S. Paramesh.

The application for the patent was published on July 03, 2026, under issue no. 27/2026.

Abstract: The present invention discloses a system and method for automated detection of suicidal ideation in social media content using machine learning and natural language processing. The system comprises a preprocessing pipeline that normalizes raw social media text by removing URLs, stopwords, punctuation, and applying lowercasing and stemming, followed by a feature extraction engine that generates TF-IDF vector representations from the cleaned text. A logistic regression classifier trained on the Kaggle Suicide Prediction Dataset receives the feature vectors and outputs a probability score indicating the likelihood of suicidal content, achieving a classification accuracy of 91.04% and a precision of approximately 92% on a held-out test set. A risk stratification engine maps the classifier output to three tiered risk categories—low risk (0–30%), moderate risk (30– 70%), and high risk (70–100%)—providing nuanced and actionable assessments for proportional crisis intervention responses. The entire pipeline is deployed as a real-time interactive web application built on the Streamlit framework, enabling mental health professionals, healthcare providers, social media platform administrators, educational institutions, and crisis helplines to rapidly assess submitted content and initiate timely intervention. The system is designed with a modular architecture to support future enhancement with deep learning models such as LSTM and BERT, multi-language NLP support, real-time API integration, automated alert notification, longitudinal sentiment tracking, and mobile platform deployment, ensuring long-term extensibility and impact in the domain of AI-assisted mental health monitoring at scale.

Disclaimer: Curated by HT Syndication.