Development of a Web-Based System for Predicting Depression and Suicide Attempt Using Ensemble Machine Learning Model - Research Paper - Volume 30 - Issue 6 - 2025

Aug 5, 2026

Depression is a mental illness that can worsen over time, especially when timely attention is not provided, and it may ultimately lead to suicide. The work proposes an ensemble learning approach using demographic data to detect depression and predict suicide attempts, then support prevention through a web-based application system. Using the “Forever Alone” dataset from Kaggle, which is imbalanced and balanced via SMOTE, multiple train/test splits are evaluated, with 80/20 performing best.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Abstract

Keywords

I. Introduction

Global impact of depression and suicide

Need for early prediction

Related work using machine learning

II. Materials and Methods

Dataset and preprocessing

Data balancing with SMOTE

Train/test splitting strategy

Ensemble model design (AdaBoost + Extra Trees)

III. Results and Discussion

Performance across evaluation metrics

Accuracy and recall for depression and suicide attempt

IV. Ethical Considerations

V. Conclusion

Olowookere T.A. Adegoke-Elijah A., Kayode A.A, Akinyele O.T Volume 30, Issue (6), 2025 Development of a Web-Based System for Predicting Depression and Suicide Attempt Using Ensemble Machine Learning Model Akinyele O.T (1*) Kayode A.A (1) Adegoke-Elijah A (1) Olowookere T.A(1) Received: 13/03/2025 Revised: 17/05/2025 Accepted: 18/05/2025 © 2025 University of Science and Technology, Aden, Yemen. This article can be distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. 2025 © ـــــــــــــــــــــــــــــــــــــــــــــــــــــــــــــــــــ 1Department of Computer Science, Redeemer’s University, Nigeria ,Nigeria *Corresponding Author’s Email: {akinyele11482, kayodeade, adegoke-elijaha, [olowookereta}@run.edu.ng](olowookereta}@run.edu.ng) 97 [https://journals.ust.edu/index.php/JST](https://journals.ust.edu/index.php/JST) Olowookere T.A. Adegoke-Elijah A., Kayode A.A, Akinyele O.T Volume 30, Issue (6), 2025 Development of a Web-Based System for Predicting Depression and Suicide Attempt Using Ensemble Machine Learning Model Akinyele O.T Department of Computer Science, Redeemer’s University, Nigeria Nigeria [akinyele11482@run.edu.ng](akinyele11482@run.edu.ng) Kayode A.A, Department of Computer Science, Redeemer’s University, Nigeria Nigeria [kayodeade@run.edu.ng](kayodeade@run.edu.ng) Adegoke-Elijah A. Department of Computer Science, Redeemer’s University, Nigeria Nigeria [adegoke-elijaha@run.edu.ng](adegoke-elijaha@run.edu.ng) Olowookere T.A Department of Computer Science, Redeemer’s University, Nigeria Nigeria [olowookereta @run.edu.ng](olowookereta @run.edu.ng) Abstract— Depression is a mental illness that can make a person’s life difficult and can eventually lead to suicide. Depressed individuals who do not receive timely attention develop worse conditions and may eventually commit suicide. Depression and suicide are becoming a global health concern which need to be adequately addressed. In this study, an ensemble learning model which make use of demographic data to detect depression and suicide attempt and also guide individuals from committing suicide through the web-based application system is proposed. The forever Alone demographic dataset which was downloaded from Kaggle online data repository was used, the dataset was imbalanced and was balanced using synthetic minority oversampling technique (SMOTE) . The dataset was split into 60/40, 70/30 and 80/20 train/test percentage split, however, the 80/20 train/test split performed best and it was used and reported in this study. The study employs an ensemble machine learning model, specifically Adaboost with Extra trees as base estimators for prediction. Adaboost enhances model performance especially in handling class imbalance leading to excellent accuracy. Results obtained reveal that Adaboost ensemble model outperformed all other machine learning algorithms across all evaluation metrics with 82.00% recall and 78.69% accuracy for depression, and 93.85% recall and 90.60% accuracy for suicide attempt respectively on the balanced dataset. The uniqueness of Adaboost in sequential weighting of misclassified instances which enhances model performance, especially in handling class imbalance thus leading to an excellent accuracy. It was therefore used for the prediction system. The study affirmed the prowess of ensemble machine learning model for predicting depression and suicide attempt. Ethical issues were also discussed in the study. Keywords—Ensemble Learning, Depression, Data split, Machine learning, Suicide attempt, Adaboost. I. INTRODUCTION It has been estimated that more than 300 million people worldwide suffer from depression, a mental condition that can make a person's life difficult and eventually lead to suicide [1] . Even though they display the symptoms through

FAQs

What problem does the proposed system address?

It targets early detection of depression and prediction of suicide attempts, aiming to reduce the risk of suicide through timely guidance via a web-based application.

Which dataset and preprocessing steps are used?

The study uses the “Forever Alone” demographic dataset from Kaggle, applies SMOTE to balance the imbalanced data, and evaluates several train/test splits, selecting the 80/20 split as best.

How does the ensemble model perform and which approach is used?

An ensemble model combining AdaBoost with Extra Trees base estimators is used; results show it outperforms other algorithms, achieving 82.00% recall and 78.69% accuracy for depression, and 93.85% recall and 90.60% accuracy for suicide attempt on the balanced dataset.

Aug 5, 2026

Upload by Seraphina