Hybrid Feature-Based Depression Prediction Model for Structured Mental Health Data Using Machine Learning
International Journal of Medical Toxicology and Forensic Medicine,
Vol. 16 (2026),
1 Dey 2026
,
Page 1-11
https://doi.org/10.22037/ijmtfm.v16.53235
Abstract
Background: Depression is a frequently occurring mental health disorder that can significantly impact life satisfaction, functioning, and social productivity. It is crucial that people at risk of depression are identified early so that they can be assessed and intervention provided early. This study aimed to develop a Hybrid Feature-Based Depression Prediction Model (HFDPM) using socio-demographic and behavioral features to automatically classify depression risk.
Methods: The proposed HFDPM included systematic data preprocessing, categorical feature encoding, feature preparation, and supervised machine-learning-based classification. We compared four machine learning algorithms: Logistic Regression, Random Forest, Support Vector Machine (SVM), and K-Nearest Neighbors (KNN). We evaluated model performance using accuracy, sensitivity, specificity, F1 score, and area under the receiver operating characteristic (ROC) curve (AUC).
Results: The HFDPM built using the Random Forest Classifier achieved the best overall classification performance, with an accuracy of 0.8678, a weighted F1 score of 0.8969, and a sensitivity of 86.25%. Logistic Regression had the highest discriminative ability with an AUC of 0.9338. Comparative results indicated that the predictive performance of the machine learning models varied, and the Random Forest-based ensemble performed best on the most important classification metrics.
Conclusion: The results demonstrate the feasibility of combining machine learning methods with socio-demographic and behavioral characteristics to predict depression risk. The plan to implement the HFDPM offers a feasible, scalable computational model for initial depression screening and for identifying those who may need additional professional evaluation.
- Depression prediction, Hybrid feature model, Machine learning (ML), Random forest, Ensemble learning (EL), Mental health analytics, Hybrid feature-based depression prediction model (HFDPM)
How to Cite
References
[1] Zulfiker MS, Kabir N, Biswas AA, Nazneen T, Uddin MS. An in-depth analysis of machine learning approaches to predict depression. Curr Res Behav Sci. 2021;2:100044. [DOI: 10.1016/j.crbeha.2021.100044]
[2] Vandana, Marriwala NK, Chaudhary D. A hybrid model for depression detection using deep learning. Measurement: Sensors. 2023;25:100587. [DOI: 10.1016/j.measen.2022.100587]
[3] Huang AA, Huang SY. Exploring depression and nutritional covariates among US adults using Shapley additive explanations. Health Sci Rep. 2023;6(10):e1635. [DOI: 10.1002/hsr2.1635]
[4] Priya A, Garg S, Tigga NP. Predicting anxiety, depression and stress in modern life using machine learning algorithms. Procedia Comput Sci. 2020;167:1258-67. [DOI: 10.1016/j.procs.2020.03.442]
[5] Squarcina L, Villa FM, Nobile M, Grisan E, Brambilla P. Deep learning for the prediction of treatment response in depression. J Affect Disord. 2021;281:618-22. [DOI: 10.1016/j.jad.2020.11.104]
[6] Wang K. Interpretable ensemble learning model with Shapley additive explanations for predicting anxiety symptoms risk in Chinese older adults with body shape index abnormality. PLoS One. 2025;20(10):e0335437. [DOI: 10.1371/journal.pone.0335437]
[7] Uddin MZ, Dysthe KK, Følstad A, Brandtzaeg PB. Deep learning for prediction of depressive symptoms in a large textual dataset. Neural Comput Appl. 2022;34:721-44. [DOI: 10.1007/s00521-021-06426-4]
[8] Islam MR, Kamal ARM, Sultana N, Islam R, Moni MA, Ulhaq A. Detecting depression using K-nearest neighbors (KNN) classification technique. In: 2018 International Conference on Computer, Communication, Chemical, Material and Electronic Engineering (IC4ME2). IEEE; 2018. p. 1-4. [DOI: 10.1109/IC4ME2.2018.8465641]
[9] Byeon H. Advances in machine learning and explainable artificial intelligence for depression prediction. Int J Adv Comput Sci Appl. 2023;14(6):520-6. [DOI: 10.14569/IJACSA.2023.0140656]
[10] Gao S, Calhoun VD, Sui J. Machine learning in major depression: from classification to treatment outcome prediction. CNS Neurosci Ther. 2018;24(11):1037-52. [DOI: 10.1111/cns.13048]
[11] Mimikou C, Kokkotis C, Tsiptsios D, Tsamakis K, Savvidou S, Modig L, et al. Explainable machine learning in the prediction of depression. Diagnostics (Basel). 2025;15(11):1412. [DOI: 10.3390/diagnostics15111412]
[12] Seal A, Bajpai R, Agnihotri J, Yazidi A, Herrera-Viedma E, Krejcar O. DeprNet: a deep convolutional neural network framework for detecting depression using EEG. IEEE Trans Instrum Meas. 2021;70:1-13. [DOI: 10.1109/TIM.2021.3053999]
[13] Lin E, Kuo PH, Liu YL, Yu YW, Yang AC, Tsai SJ. A deep learning approach for predicting antidepressant response in major depression using clinical and genetic biomarkers. Front Psychiatry. 2018;9:290. [DOI: 10.3389/fpsyt.2018.00290]
[14] Nickson D, Meyer C, Walasek L, Toro C. Prediction and diagnosis of depression using machine learning with electronic health records data: a systematic review. BMC Med Inform Decis Mak. 2023;23(1):271. [DOI: 10.1186/s12911-023-02341-x]
[15] Shrive FM, Stuart H, Quan H, Ghali WA. Dealing with missing data in a multi-question depression scale: a comparison of imputation methods. BMC Med Res Methodol. 2006;6:57. [DOI: 10.1186/1471-2288-6-57]
[16] Ndaba S. Class imbalance handling techniques used in depression prediction and detection. Int J Data Min Knowl Manag Process. 2023;13(1/2):17-33. [DOI: 10.5121/ijdkp.2023.13202]
[17] Huo Z, Qian X, Huang S, Wang Z, Mortazavi BJ. Density-aware personalized training for risk prediction in imbalanced medical data. Proc Mach Learn Res. 2022;182:101-22. [Link]
[18] Masum M. Comprehensive Depression Dataset for ML Research [dataset]. Kaggle; 2024. [Link]
[19] Masum M. Comprehensive Depression Dataset for ML Research [dataset]. Kaggle; 2024 [cited 2026 Sep 24]. [Link]
- Abstract Viewed: 45 times
- 53235 Downloaded: 29 times