PREDICTION OF DEPRESSION USING MACHINE LEARNING TOOLS TAKING CONSIDERATION OF OVERSAMPLING

Depression is a psychiatric condition characterized by persistent sense of sadness and dullness. It also known as severe burdensome problem or clinical sorrow, it impacts how person feels, thinks, behaves triggers slew emotional physical issues. Various components are liable for this issue, many related sicknesses expanding because infection. not just at risk well-being perils, yet produces perilous social offense, like self-destruction family misuse. In study, we used machine learning methods such Random Forest (RF), Logistic Regression (LR), Naive Bayes (NB). We accuracy, precision, recall, F1-score to survey the exhibition assessment arrangement results. These algorithms developed analyzed Confusion matrices through data augmentation assess classification performance. This study technologies predict depression revealed significance trait. Then have tried utilize an oversampling technique that shows distinction in model execution. Indeed, wanted see well recommended performed before after rebalancing standardized data. our suggested framework, RF classifier better with 89% accuracy 90% precision than other models.