Leveraging Nlp for Mental Health Detection: a Machine Learning Approach to Suicide Risk Classification

Psychiatric disorders such as depression, panic, and others continue to be a significant global issue that requires prompt identification and intervention approaches. In this research, machine learning methods are utilized to categorize suicide-related textual data, drawing from a dataset obtained from online platforms. Data preprocessing is performed through text cleaning and TF-IDF vectorization before applying five machine learning models-logistic regression (LR), naïve bayes (NB), random forest (RF), support vector machine (SVM), and XGBoost-for classification. The results emphasize the efficiency of traditional linear models in analyzing mental health data for text classification tasks. This study adds to the field of suicide risk detection through artificial intelligence, offering valuable insights into computational methods for identifying early signs of mental health issues.