Machine Learning Algorithms Applied with Questionnaire Dataset to Investigate Anxiety and Depression

It is discovered that machine learning approaches and artificial intelligence are extensively examined in a variety of research articles when it comes to the identification of psychologically linked emotional activities. Studies employing datasets like face pictures, personal videos, audio, questionnaire-response, etc., demonstrated the usefulness of artificial intelligence and machine learning methods for the identification of human emotions like anger, happiness, sadness, etc. Human emotions are the result of psychological processes that may be impacted by external everyday routines. The questionnaire-based dataset used in the proposed study shows how anxiety and depression symptoms are configured. The DASS-21 questionnaire dataset includes a common set of questions on everyday activities. The experiment collects participants’ answers to the questions and then uses machine learning techniques to analyze each response. Textual data is used to compile responses from people. The features from the gathered responses are extracted using machine learning techniques. The model uses textual preprocessing units for the textual response dataset in order to more accurately categorize anxiety and depressive symptoms. The written responses have been translated into relevant portions that provide features for this module. The DASS-21 questionnaire-based dataset is subjected to the application of five distinct machine learning techniques in the current method. The model can examine characteristics against large datasets of complex text. The model does feature extraction and classification using methods including SVM, decision trees, random forests, and Naive Bayes. The model effectively compares the use of several algorithms for categorizing anxiety and depressive symptoms.