Mental health among university students in Bangladesh forecasted using machine learning

Abstract

Medical professionals diagnose depression based on patient self-reporting and mental status questionnaires. Mentally ill people are reluctant to seek mental health treatment, in addition to the fact that the approaches vary greatly depending on the patient's state of mind at the time. Academic fields typically offer their students bright futures. But among the many other things that could cause a student to experience depression are peer pressure, academic competition, loneliness, and many other things. This study aims to identify depression in college students using a big data analytics template. It is asserted once more that the framework models the relationship between depression and factors like isolation and separation, which are thought to have the most profound effect on students. In summary, the journal assesses how well the suggested framework performs using a sizable real dataset gathered from various Bangladeshi university students, and it demonstrates that machine learning models detect depression in universities more accurately than traditional methods. I will apply six pre-trained models (Logistic Regression, Decision Tree, Random Forest, Gradient Boost, KNN, and Naive Bayes) to classify and identify this issue. Among them, Logistic Regressiongave performed better than any other proposed model. In comparison to other models, the model provided good detection accuracy.