Mental Health Risk Detection in Text Using NLP and Logistic Regression

"Mental Health Risk Detection in Text Using NLP and Logistic Regression"

Balla Abhishek vincent Dept of ECE IARE

Dr. S China Venkateshwarlu Professor Dept of ECE IARE

Dr. V Siva Nagaraju Professor Dept of ECE IARE

Abstract -  This project presents a machine learning approach to detect mental health risks from user-generated text using Natural Language Processing (NLP) techniques. A dataset containing labeled mental health-related text is preprocessed with stopword removal, lowercasing, and token cleaning. The clean text is then vectorized using TF-IDF and classified using a Logistic Regression model. The model achieves strong performance metrics, including accuracy and precision, in identifying potential mental health concerns. A prediction function is also developed for real-time risk assessment of new input text. This tool aims to assist in early detection and intervention by flagging potential indicators of mental distress.

Key Words:  Mental Health, Natural Language Processing (NLP), Text Classification, Logistic Regression, TF-IDF Vectorization.