Development of a Student Depression Prediction Model Based on Machine Learning with Algorithm Performance Evaluation

Aug 5, 2026

Upload by Quinn

This research explores the use of machine learning to predict depression among university students, leveraging a dataset of 2,028 responses with PHQ-9 scores and academic-demographic attributes. A structured modeling workflow is applied, including feature selection and normalization. Model effectiveness is assessed using accuracy, precision, recall, and F1-score. After hyperparameter tuning, the support vector machine (SVM) model’s accuracy increases from 58.8% to 99.5%. The results support a proactive identification framework for collegiate mental well-being monitoring and future real-time deployment with expanded digital counseling and behavioral analytics.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages23

Reading Time58 min

AccessFree to download

Introduction

Motivation and mental health context

Role of machine learning in early detection

Research gaps and limitations

Proposed comparative framework

Abstract

Journal of Information Systems and Informatics Vol. 7, No. 2, June 2025 e-ISSN: 2656-4882 p-ISSN: 2656-5935 DOI: 10.51519/journalisi.v7i2.1087 Published By DRPM-UBD Development of a Student Depression Prediction Model Based on Machine Learning with Algorithm Performance Evaluation Penni Wintasari Simarmata1, Putri Taqwa Prasetyaningrum2 1,2Information System, Mercu Buana University, Yogyakarta, Indonesia [Email:](Email:1 penniwitasari1122@gmail.com)[1](Email:1 penniwitasari1122@gmail.com)[ penniwitasari1122@gmail.com](Email:1 penniwitasari1122@gmail.com), [2](2 putri@mercubuana-yogya.ac.id)[ putri@mercubuana-yogya.ac.id](2 putri@mercubuana-yogya.ac.id) Abstract This research explores the implementation of machine learning to predict depression among university students using a dataset of 2 .028 responses containing PHQ-9 scores and academic-demographic attributes. The research implements a structured modeling process involving feature selection, normalization, the model’s efficacy was gauged through a suite of evaluate measures, encompassing accuracy, precision, recall, F1-score, The support vector machine (SVM) model’s accuracy improved from 58.8% to 99.5% after hyperparameter tuning. This investigation lends itself to the advancement of a proactive identification framework, which hold potential for incorporation within collegiate mental well-being surveillance infrastructures. Future implementations may consider real-time models and expand data sources through digital counseling systems and behavioral analytics Keywords: Classification Algorithm, Depression Prediction, Machine Learning, Model Development, Model Evaluation 1. INTRODUCTION Psychological conditions, particularly those affecting students, have become a growing concern, gaining significant attention in recent years due to the pressures they face—academic demands, social expectations, and uncertainty about the future [1] . Among these, depression stands out as one of the most prevalent psychological disorders. This condition is often marked by persistent low mood, prolonged feelings of sadness, hopelessness, guilt, and worthlessness, which severely impact one’s emotional well-being [2] . According to the World Health Organization (WHO), approximately 280 million people worldwide were living with depressive disorders in 2019, with a disturbing rise in prevalence among adolescents and young adults. A study revealed that 29% of university students suffer from anxiety disorders, and 25% experience depression, with varying degrees of severity, from mild to severe [3] . These mental health challenges significantly influence various aspects of cognition, emotion, and behaviour, 1283 This work is licensed under a Creative Commons Attribution 4.0 International License. p-ISSN: 2656-5935 [http://journal-isi.org/index.php/isi](http://journal-isi.org/index.php/isi) e-ISSN: 2656-4882 ultimately diminishing motivation to engage in daily activities or maintain a social life [4] . Despite the growing mental health crisis, many students hesitate to seek professional help, largely due to the stigma surrounding mental health issues. The fear of being labelled as "weak" or incapable often prevents them from accessing the psychological services they need [5] . In addition, the lack of mental health awareness and the shortage of professional resources further complicate efforts to provide timely and effective support, hindering early intervention. In light of these challenges, recent advancements in Artificial Intelligence (AI), particularly in Machine Learning (ML), have presented new opportunities for detecting and diagnosing mental health issues. ML algorithms can analyse diverse data types—such as surveys, behavioural patterns, and psychological symptoms—offering a more objective, efficient, and scalable approach to mental health detection [6][7] . Prior research has demonstrated the potential of ML-based depression prediction models as effective t

FAQs

What dataset and target measure are used for depression prediction?

The study uses a dataset of 2,028 responses containing PHQ-9 scores along with academic-demographic attributes to model student depression.

Which algorithms are compared in the proposed framework?

Six machine learning algorithms are evaluated: SVM, Logistic Regression, K-Nearest Neighbors, Random Forest, Decision Tree, and Naive Bayes.

How does hyperparameter tuning affect the SVM model performance?

With hyperparameter tuning, the SVM accuracy improves from 58.8% to 99.5%, indicating a substantial performance gain.

Aug 5, 2026

Upload by Quinn