A Comparative Evaluation of Machine Learning and Deep Learning Models for Detecting Depressive behavior on X

Early detection of mental health concerns through social media platforms like X offers significant potential for proactive intervention. However, identifying nuanced emotional and psychological distress from short, informal texts remains a complex challenge. This study aims to bridge this gap by developing and comprehensively evaluating a spectrum of machine learning, deep learning, and transformer-based models for detecting depressive behavior from X posts. A robust natural language processing (NLP) pipeline was established, encompassing data cleaning, text preprocessing, feature extraction, and classification. Four distinct architectures were investigated: Logistic Regression, Linear Support Vector Machine (SVM), CNN-LSTM, and Fine-Tuned Distil BERT. Traditional models utilized TF-IDF representations, while deep learning and transformer models employed token-based representations. Model performance was rigorously benchmarked using accuracy, precision, recall, and F1-score. Experimental results reveal that all models achieved high efficacy with accuracies exceeding 96%. Notably, the Fine-Tuned Distil BERT secured the best overall performance with an accuracy of 97.46% and an F1-score of 96.68%, whereas Linear SVM demonstrated exceptional sensitivity with the highest recall of 97.05%. These findings demonstrate that transformer-based models and advanced NLP pipelines significantly enhance the reliability of automated mental health screening on social media, providing a scalable foundation for intelligent, text-based psychological monitoring systems.

Depression Detection, Natural Language Processing (NLP), Transformer Models, Machine Learning

1. Samsudin, S., Ismail, R., Mohd Daud, S. N., & Yahya, S. (2024). The prevalence and underlying factors of mental health disorders in youth. Journal of Affective Disorders Reports, 15, 100680. [Google Scholar] [Crossref]

2. Russell, S., & Norvig, P. (2021). Artificial Intelligence: A Modern Approach (4th ed.). Pearson. [Google Scholar] [Crossref]

3. Biswas, S., & Hasija, Y. (2022). Predicting depression through social media. Lecture Notes on Data Engineering and Communications Technologies, 112, 45-58. [Google Scholar] [Crossref]

4. Cavazos-Rehg, P. A., Krauss, M. J., Sowles, S., Connolly, S., Rosas, C., Miller, M. B., & Bierut, L. J. (2016). A content analysis of depression-related tweets. Computers in Human Behavior, 54, 351-357. [Google Scholar] [Crossref]

5. Gu, L., Li, M., & Li, Y. (2026). Linguistic markers of depression and emergent self-stigma in online self-disclosures: A mixed-methods study on Chinese social media. Journal of Affective Disorders, 395, 112-123. [Google Scholar] [Crossref]

6. Adugna, T. D., Ramu, A., & Haldorai, A. (2024). A review of pattern recognition and machine learning. Journal of Machine and Computing, 4(1), 12-25. [Google Scholar] [Crossref]

7. Chatterjee, M., Kumar, P., Samanta, P., & Sarkar, D. (2022). Suicide ideation detection from online social media: A multi-modal feature based technique. International Journal of Information Management Data Insights, 2(2), 100095. [Google Scholar] [Crossref]

8. Ahmed, M., Ansari, M., Qadir, A., & Hashmi, D. (2025). Depression detection using machine learning techniques on X/Twitter data. International Journal of Information Technology and Computer Engineering, 13, 225–232. [Google Scholar] [Crossref]

9. Alphonsa Sini, P. J., & Sherly, K. K. (2023). Early detection of anxiety, depression and stress among potential patients using machine learning and deep learning models. Proceedings of the 2nd International Conference on Computational Systems and Communication (ICCSC), 104-110. [Google Scholar] [Crossref]

10. Chiong, R., Budhi, G. S., Dhakal, S., & Cambria, E. (2021). Combining sentiment lexicons and content-based features for depression detection. IEEE Intelligent Systems, 36(6), 99-105. [Google Scholar] [Crossref]

11. Amanat, A., Rizwan, M., Javed, A. R., Abdelhaq, M., Alsaqour, R., Pandya, S., & Sujatha, R. (2022). Deep learning for depression detection from textual data. Electronics (Switzerland), 11(5), 780. [Google Scholar] [Crossref]

12. Prabhakar, K., & Kavitha, V. (2024). An evolutionary approach for depression detection from Twitter big data using a novel deep learning model with attention based feature learning mechanism. Automatika, 65(2), 512-524. [Google Scholar] [Crossref]

13. Tejaswini, V., Babu, K. S., & Sahoo, B. (2024). Depression detection from social media text analysis using natural language processing techniques and hybrid deep learning model. ACM Transactions on Asian and Low-Resource Language Information Processing, 23(1), 1-22. [Google Scholar] [Crossref]

14. Patkar, U. C., Sharma, R., & Singh, K. (2024). Object and sentiment pattern recognition using machine learning and deep learning. International Journal of Intelligent Systems and Applications in Engineering, 12(1s), 415-422. [Google Scholar] [Crossref]

15. Turuba, R., Johnson, P., & Smith, L. (2025). A summative content analysis of mental health-related content on social media platforms. Digital Health, 11, 1-14. [Google Scholar] [Crossref]

16. Angskun, T., Tipprasert, S., Thippongtorn, A., & Angskun, J. (2024). D₂X: Depression detection system through X using hybrid machine learning. IEEE Access, 12, 172820–172831. [Google Scholar] [Crossref]

17. Tejaswini, V., Babu, K. S., & Sahoo, B. (2024). Depression detection from social media text analysis using natural language processing techniques and hybrid deep learning model. ACM Transactions on Asian and Low-Resource Language Information Processing, 23(1), 1-22. [Google Scholar] [Crossref]

• From Anxiety to Artistry: Evaluating the Impact of a Specialised Guest Lecturer Webinar on Students' Perceived Paraphrasing Competencies
• “Bro”, “Bestie”, and “Slay”: Slang and Digital Identity Among Generation Z WhatsApp Communication
• Oral Reading Performance and Language Subject Achievement among Grade 3 Learners