Identification and comparison of factors behind mental turbulence of public and private university students

Abstract

Among the young generation of Bangladesh, especially university students, mental health is a challenging but ignored topic. The primary goal of this paper is to analyze mental states found in university students all while trying to find out comparative distinction between factors prominent in public and private university environment. Professional consultation has been taken to determine probable factors related to a student’s university life and mental health. The dataset is split into public and private university students to map the specific involvement of aforementioned factors. SMOTE has been used to handle data imbalance for minority classes. Machine learning algorithms such as Logistic Regression, K-Nearest Neighbour Regression and Random Forest Regression have been evaluated with MAE and R2. The most optimal regression algorithm followed by Apriori data mining algorithm have been discussed to build a hybrid model. Most influential factors are identified, comparative analysis of selected features’ influence levels and associations are explored and probable causes are discussed.