Detection of Mental Well-Being Status Through Data-Driven Approaches

Mental well-being disorders are among the most significant challenges in the modern lifestyles, and it is well-established that early detection of diseases is essential for effective prevention. On the other hand, Machine Learning (ML) algorithms currently play a valuable role in disease detection. The aim of this study is not only to investigate the performance of various ML classifiers but also to propose a modern technique for mental health diagnosis. In this context, our research considers Bootstrap Aggregating (Bagging), Extremely Randomized Trees (ExtraTrees), Passive-Aggressive, Light Gradient Boosting Machine (LGBM), Perceptron, and Stochastic Gradient Descent (SGD) algorithms, which are among the widely recognized ML classifiers in literature. To address the factors contributing to mental health illnesses among the selected individuals, we employ a three-phase data processing approach: segmentation, feature extraction, and classification. Analyzing feature importance from the selected dataset, our study highlights the significant impact of age, family history, and workplace environment on a worker's mental health status.