In the research, different machine learning methods were tested to predict mental health issues using survey data. The algorithms tried to see how accurately they could classify people's mental health status. Gradient Boosting turned out to be the best, with an accuracy of 88.80%, showing the highest ability to identify mental health problems from the survey answers. Neural Networks came next with 88.00% accuracy, followed by Extreme Gradient Boosting and Deep Neural Networks. The combination of all these methods also did quite well, achieving an accuracy of 85.60%. Overall, the results suggest that these machine learning techniques can predict mental health issues with over 80% accuracy, offering promise for automated diagnosis by mental health professionals.