Related Experiment Video
Updated: Jun 24, 2025
06:37
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
Published on: December 15, 2023
2.7K
Mental health analysis of international students using machine learning techniques
Muhammad Azizur Rahman1, Tripti Kohli1
1Cardiff School of Technologies, Cardiff Metropolitan University, Llandaff Campus, Cardiff, Wales, United Kingdom.
Plos One
|June 6, 2024
Summary
International students face mental health challenges abroad. This study used machine learning to predict depression, identifying key demographic and psychosocial factors, with Random Forest achieving 80% accuracy.
Area of Science:
- Psychology
- Computer Science
- Sociology
Background:
- International student mental health is a growing concern due to challenges like cultural differences and homesickness.
- Limited research exists on factors influencing international students' psychological well-being and the application of machine learning in this domain.
Purpose of the Study:
- To analyze demographic, cultural, and psychosocial factors impacting international students' mental health.
- To develop a machine learning model for predicting depression among international students in the United Kingdom.
Main Methods:
- Utilized primary survey data and secondary data from 'A Dataset of Students' Mental Health and Help-Seeking Behaviors'.
- Applied four machine learning models (Logistic Regression, Decision Tree, Random Forest, K Nearest Neighbor) to a training/testing split dataset.
- Evaluated model performance using Accuracy, Sensitivity, Specificity, Precision, and AU-ROC curve.
Main Results:
- Identified significant demographic variables (loan status, gender, age, marital status) and psychosocial factors (financial difficulties, academic stress, homesickness, loneliness).
- The Random Forest model achieved the highest prediction accuracy at 80% for depression.
Conclusions:
- Demographic and psychosocial factors significantly influence international students' mental health.
- Machine learning, particularly Random Forest, shows promise in predicting depression among this population.
