A Machine Learning Approach for Identifying Mental Health Disorders and Depression in Psychoactive Drug Users

Mental illnesses are among the most prevalent problems worldwide regardless of age. The current investigation examines ML (Machine Learning) methods to predict Depression and CMD (Common Mental Disorders) for people who utilize Psychoactive Substances. An assortment of techniques such as Support Vector Machine(SVM), Random Forest(RF), K-Nearest Neighbour(KNN), and Naive Bayes, as well as a voting classifier that combines Baging Classifier, along with a Decision Tree, are used. Utilizing Sequential Backward Selection (SBS), the findings demonstrated exceptional accuracy, with Support Vector Machine (SVM) achieving 84.33% for Common Mental Disorder (CMD) predictions as well as 83.93 % for Depression. In an attempt to achieve higher accuracy, possibly exceeding 90%, ensemble techniques such as the voting classifier were investigated.