A person is said to be suffering from a mental disorder if their cognition, emotion control, or behavior are all affected. Despite the prevalence of mental illness, it is often under diagnosed. There is a lot of curiosity about the link between depression and social media. Individuals' well-being is negatively impacted by social media use, according to several experts. Many attempts have been made to analyze individual postings using Machine Learning (ML) methods to identify mental patients on social media. Datasets taken from different social media platforms and the sorts of characteristics used in the identification of mental disorders are discussed in this study. Research methods included decision trees, random forests, support vector machines, recurrent neural networks, convolutional neural networks as well as logistic regression. In comparison to the other methods, the Convolutional Neural Network, also known as (CNN), achieved the best accuracy, which was 91.08 percent, for the diagnosis of mental conditions. Accuracy rates of 85.87 percent, 81.22 percent, and 85 percent were achieved by using other methods such as random forests, Support Vector Machines (SVM), and recurrent neural networks (RNN), respectively. It has been shown that ML techniques, when useful to text data collected from users of social media platforms, can be an effective method for detecting depression and can one day serve as supplementary tools in the field of public mental health.