STUDENT MENTAL HEALTH PREDICTIVE ANALYSIS Investigation towards student mentality in undergraduate education Mohammed Abdulaziz Maki Department of Information System University of Bahrain 20197476@stu.uob.edu.bh Mohammed Essam Alrayes Department of Information System University of Bahrain 20191802@stu.uob.edu.bh Mustafa Mohammed Ahmed Department of Infomaaion System University of Bahrain 20192808@stu.uob.edu.bh Younis Hesham Alaradi Department of Information System University of Bahrain 20192885@stu.uob.edu.bh Osama Mohammed Shakeel Department of Information System University of Bahrian 20191465@stu.uob.edu.bh Abstract— This research paper focuses on the mental health of university students and aims to address the lack of awareness and mistreatment of mental disorders among this population. By utilizing big data analytics and machine learning techniques, the study explores the factors and variables that contribute to depression, one of the most prevalent mental health issues among students. The project's goal is to develop and implement predictive models to detect early signs of depression, enabling timely intervention and assistance. By tailoring support systems to individual needs, educational institutions can effectively allocate resources and promote student well-being and academic achievement. The research also aims to guide the creation of preventive programs and policies to foster a loving and inclusive school environment. R Studio and Weka are employed as primary tools for data analysis, utilizing decision trees and conducting further analyses, as necessary. Based on existing research, key elements influencing students' mental health include gender, age, major, GPA, study circumstances, family status, and social factors. The literature review confirms that university students face prominent levels of pressure, stress, and other factors associated with higher education degrees, which contribute to the prevalence of mental disorders. The analysis of a sample of 219 students from India, drawn from the "Entrepreneurial Competency in University Students" online dataset, reveals that business and arts students are particularly susceptible to mental disorders compared to other disciplines. Additionally, the study confirms that female students are highly likely to be affected by mental disorders. The findings of this research contribute to early detection, intervention, and the development of tailored support systems in educational settings. They also inform the design of preventive programs and policies aimed at creating a nurturing and inclusive school environment, ultimately promoting the well-being and academic success of students. Keywords—Student, Mentality, Random Forest, undergraduate, education, prediction, depression. I. INTRODUCTION Students' mental health is frequently disregarded or mistreated; even students may not appear to comprehend that they may have mental disorders, either due to a lack of awareness of the symptoms or due to other causes relating to their lives or education. This is a matter that should not be taken lightly whatsoever. By using big data analytics, students can find patterns or causes to mental health. The goal of this project is to use machine learning techniques and models to study and analyze some of the factors and variables that may be the cause of one of the most common mental health issues among students (Depression), and then implement those models to predict if the student may suffer from Depression. Early detection provides for immediate action and assistance, lowering the probability of more serious problems later on. It can also assist schools in tailoring their support systems to individual requirements, ensuring that resources are efficiently spent. Furthermore, such research can help to guide the creation of preventative programs and policies aimed at building a more loving and inclusive school environment, eventually promoting kids' well-being and academic achievement. In this project, we opted to utilize R Studio and Weka as our primary machine learning tools to analyze the data. We will use Weka for decision trees and R Studio to display the data and do further analyses as needed. According to research on student mental health conducted by [1], the following elements have the greatest influence on students' mental health: gender, age, major, GPA and study circumstances, family status, and social factors. II. LITERATURE REVIEW University students are often exposed to mental disorder problems due to high pressure circulated by high focus, stress, and other factors surrounding higher education degrees. This was confirmed by passing sample of 219 students from India as shown in the online dataset entitled “Entrepreneurial Competency in University Students” with a sum of sixteen different variables. The analysis was accomplished through the application of three main algorithms such as K-Nearest Neighbor (KNN); Support Vector Machine (SVM); Decision trees (DT), as well as the elimination of some variables that do not lead to the intended variable which articulates that business and arts students are highly exposed to be affected by mental disorder in comparison with other disciplines. On the other side, it confirmed that female students are highly likely to be impacted by mental disorders [2].