Objective: The study aimed to assess the reliability and validity of the Psychological factors, namely Depression, Anxiety, and Stress Scale-21 (DASS- 21), among University students in Delhi. Methods: The DASS-21 questionnaire was administered by conducting a survey where around 100 samples were randomly selected. A comparison and training model was formed using the benchmark dataset along with the original data collected in the study. Three supervised machine learning models were trained on the same. The best model was selected and tested on the originally collected data.Conclusion: The factors selected using machine learning techniques affect an individual's severity. To further verify these factors, practitioners were engaged to identify the specific features that influence these psychological parameters. These results helped to understand the importance and use of machine learning techniques for analyzing the severity of stress, anxiety and depression scales amongst individuals. The testing accuracy achieved was similar to the training accuracy indicating the model did not have any anomaly and could be used for predicting the severity of stress, anxiety and depression among university students in India.

Keywords : Stress, Anxiety, Depression, Correlation, Supervised Machine Learning, Testing Accuracy.