A Machine Learning Approach to Classify Depression and Anxiety

A person's mood and overall health are significantly affected by psychological disorders like anxiety and depression. Since anxiety and depression have many symptoms in shared, distinguishing between them is the main issue. By employing a dataset of numerical replies, this work seeks to solve this problem by pinpointing the most important characteristics necessary for accurate classification. The dataset comprises 4184 individuals diagnosed by the French university and focuses on people aged 18 and over. SMOTE, which stands for Synthetic Minority Over-sampling Technique, has been used to solve the problem of class imbalance, as there were more people in the study population with no disorder. Random Forest feature selection found these 10 most relevant attributes to symptoms of anxiety and depression. The classification then proceeded with the supervised machine learning model, K-Nearest Neighbors (KNN), Random Forest, and Support Vector Machine (SVM). This was done by applying the 10-fold cross-validation method to achieve consistent and trustworthy results. Random Forest showed the highest performance among the others, thus being the best method to differentiate between depression and anxiety.