A Comparative analysis of predictive machine learning models on student mental health impact on academic outcomes

Brajen Kumar Deka; Amit Dutta; Kshirod Sarmah; Nayan Kumar Sarkar; Swapnanil Gogoi; Juniorika Lyngdoh; Sisir Kumar Rajbongshi; Ajit Das

doi:10.29284/24zhsa20

Authors

Brajen Kumar Deka Department of Computer Applications, Assam Don Bosco University, Guwahati, India
Amit Dutta Department of Computer Applications, Assam Don Bosco University, Guwahati, India
Kshirod Sarmah Department of Computer Science, Pandit Deendayal Upadhyaya Adarsha Mahavidyalaya, Goalpara, India
Nayan Kumar Sarkar Department of Artificial Intelligence and Machine Learning, IILM University, New Delhi, India
Swapnanil Gogoi Department of Computer Science & Information Technology, GUCDOE, Gauhati University, Guwahati, India
Juniorika Lyngdoh Department of Computer Applications, Assam Don Bosco University, Guwahati, India
Sisir Kumar Rajbongshi Department of Computer Science, Pandit Deendayal Upadhyaya Adarsha Mahavidyalaya, Goalpara, India
Ajit Das Department of Computer Science and Technology, Bodoland University, Kokrajhar, India

DOI:

https://doi.org/10.29284/24zhsa20

Keywords:

Anxiety, College Students, Depression, Gender Differences, Mental Health, Machine Learning, Predictive Patterns

Abstract

Mental health (MH) profoundly influences emotional, psychological, and social well-being throughout all life stages, impacting how individuals think, feel, and cope with stress. This study explores the relationship between academic life and MH among college students via a survey comprising 15 targeted questions addressing depressive symptoms. Most responses were binary, with some scaled to reflect symptom frequency. The analysis assessed question-wise response accuracy, revealing differences between boys’ and girls’ students “Yes” and “No” answers. Using five machine learning classifiers – Logistic Regression (LR), Support Vector Machines (SVM), K-Nearest Neighbors (KNN), Decision Trees (DT), and Naïve Bayes (NB) – the study achieved notable prediction accuracies, with Logistic Regression attaining 92.4% accuracy for depression detection and SVM achieving 94.1% accuracy for anxiety recognition. These results underscore the feasibility of employing predictive models for early identification of MH issues in academic settings. This work highlights the importance of gender-specific insights in MH assessment, contributing to the development of targeted interventions aimed at improving student well-being and academic success.

References

[1] Alharthi H. Predicting the level of generalized anxiety disorder of the coronavirus pandemic among college students using artificial intelligence technology. Proceedings of the 9th Distributed Computing and Applications for Business Engineering and Science. 2020; 218-221.

http://doi.101.1109/DCABES60732.2020.00064

[2] Cheung DK, Tam DKY, Tsang MH, Zhang DLW, Lit DSW. Depression, anxiety and stress in different subgroups of first-year university students from 4-year cohort data. Journal of Affective Disorders. 2020; 274:305-314. http://doi.10.1016/JJAD.2020.05.041

[3] Hou Y, Xu J, Huang Y, Ma X. A big data application to predict depression in the university based on the reading habits. Conference: 2016 3rd International Conference on Systems and Informatics (ICSAI 2016). 2017; 1085-1089. http://doi.10.1109/ICSAI.2016.7811112

[4] Hasanbasic A, Spahic M, Bosnjic D, Adzic HH, Mesic V, Jahic O. Recognition of stress levels among students with wearable sensors. Conference: 2018 18th International Symposium INFOTEH-JAHORINA (INFOTEH). 2019. http://doi.10.1109/INFOTEH.2019.8717754

[5] Choudhury AA, Khan MRH, Nahim NZ, Tulon SR, Islam S, Chakrabarty A. Predicting depression in Bangladesh undergraduates using machine learning. In proceedings of the 2019 IEEE Region 10 Symposium (TENSYMP 2019). 2019; 789-794. http://doi.10.1109/TENSYMP46218.2019.8971369

[6] Sau A, Bhakta I. Predicting anxiety and depression in elderly patients using machine learning technology. Healthcare Technology Letters. 2017; 4(6):238-243.

[7] Sharif M, Liu J, Ren W. Impact of COVID-19 pandemic on micro, small, and medium-sized enterprises in Pakistan. Research in Globalization. 2020; 2:100018. http://doi.10.1016/J.REDGLO.2020.100018

[8] Harikrishnan U, Ali A, Sobhana H. Assessment of mental health status among school-going adolescents in North East India: a cross-sectional school-based survey. Asian Journal of Psychiatry. 2017; 30 (114-117). http://doi.org/10.1016/j.ajp.2017.08.021

[9] Chen X, Genc Y. A systematic review of artificial intelligence and mental health in the context of social media. Proceedings of the Artificial Intelligence in HCL; Virtual. 2022; pp. 353-368. [Google Scholar]

[10] Nadia OE, Juman MA, Reem BB, Danish AA, Muhammed AK. Smartphone addiction, stress, and depression among university students. Clinical Epidemiology and Global Health. 2023; 25(3):101487. http://doi.org/10.1016/j.cegh.2023.101487

[11] Haruna U, Mohammed AR, Braimah M. Understanding the burden of depression, anxiety and stress among first-year undergraduate students. 2025; 25(1): 1-9. http://doi.org/10.1186/s12888-025-07069-8

[12] Costantini L, Pasquarella C, Odone A, Colucci ME, Costanza A, Serafini G, et al. Screening for depression in primary care with patient health Questionnaire-9 (PHQ-9): a systematic review. J. Affect Disord. 2021; 279, 473-483.http://doi.org/10.1016/j.jad.2020.09.131

[13] Al-Mamun F, Islam J, Muhit M, Mamun MA. Prevalence of emotional and behavioral problems among adolescents in Bangladesh. Soc. Psychiatry Epidemiology. 2024; 59(12): 2215-2225.

http://doi.org/10.1007/s00127-024-02673-7

[14] Cherian AV, Armstrong G, Ravindra RM. Mental health, suicidality, health and social indicators among college students across nine states in India. Indian Journal of Psychological Medicine.

https://doi.org/10.1177/02537176241244775

[15] Agarwal P, Mundaragi RS, Kohad RS, Allada R, and Bharadwaj SR. Predicting student depression using machine learning (Decision Tree, Logistic Regression, Random Forest). International Journal of Innovative Science and Research Technology. 2025; 10(1): 940-945. https://doi.org/10.5281/zenodo.14737958

[16] Nuarini S, Fauziah S, Mayangky NA, and Nurfalah R. Comparison algorithm on machine learning for student mental health data — KNN among models used to predict student mental health conditions during and after the pandemic. Journal of Medical Informatics Technology (MedInfoTech). 2023;1(3): 81-85.

https://doi.org/10.37034/medinftech.v1i3.18

[17] Zhang L. A logistic regression modeling study of college students’ mental health status in predicting suicidal behavior. Applied Mathematics and Nonlinear Sciences. 2024; 9(1):1-14.

https://doi.org/10.2478/amns-2024-2837

[18] Torres Cruz F, Coaquira Flores EE, Condori Quispe SJ. Prediction of Depression Status in College Students Using a Naïve Bayes Classifier Based Machine Learning Model. 2023; 1-6.

https://doi.org/10.48550/arXiv.2307.14371

[19] Wijaya V, Rachmat N.Comparison of SVM, Random Forest, and Logistic Regression performance in student mental health screening. Journal of Electrical Engineering and Computer Sciences (JEECS). 2024; 9(2): 173-184.

[20] Prithivi R, Patil J. Mental health laws in India: A critical analysis. International Journal of Health Science. 2022; 6(S1):9775-9786. https://doi.org/10.53730/ijhs.v6nS1.7288

[21] George DM. An analysis of legal provisions and regulatory mechanisms in ensuring student well-being and mental well-being. Indian Journal of Legal Review. 2024; 4(1):324-329.

[22] McClenaghan E. Spearman Rank Correlation. Technology Network, 2023.

https://www.technologynetworks.com/tn/articles/spearman-rank-correlation-385744