Mental Health Prediction in Work Place

Abstract

This paper presents Employee mental health plays a vital role in sustaining organizational performance and overall workplace harmony. Conventional methods of assessing stress and related issues often depend on surveys or self-disclosure, which are limited by bias, hesitation, and lack of immediacy. To overcome these gaps, this work introduces a web-based mental health prediction framework powered by machine learning. The system collects structured employee inputs through a digital form, processes the data, and employs a Random Forest model to classify individuals into Low, Medium, or High risk categories. Additionally, textual mood analysis is integrated to improve the reliability of outcomes. Based on the results, personalized recommendations are automatically generated for employees to encourage wellness practices. An HR-facing dashboard visualizes organizational trends with charts and gender-based insights, ensuring that collective patterns are clear while individual confidentiality is preserved. Identity protection features and secure admin access strengthen trust and privacy within the system. Experimental validation highlights the model’s ability to balance prediction accuracy, usability, and data safety. This framework demonstrates the potential of combining artificial intelligence with workplace wellness strategies to promote healthier, more productive environments.