Using Deep Learning Techniques to Identify Depression in Social Media Interactions

Abstract

Depression is one of the most common mental problems and an important cause of suicide. The increased use of social media platforms has enabled early diagnosis of depression through the sentences expressed by users using everyday language. As social media platforms continue to play a central role in individuals' daily lives, there is a growing interest in using these platforms for mental health analysis. In this study, Twitter (nowadays X) was used to categorize depression. The aim of this study is to identify tweets with depressive characteristics among tweets retrieved from Twitter. We present an innovative model for depression prediction using a Bi-directional Long Short-Term Memory (Bi-LSTM) architecture. This model utilizes appropriate cleaning and preprocessing techniques for more accurate detection of depression using language features in tweets. For the study, a special dataset obtained through Twitter API was created and analyses were performed on this dataset. The proposed Bi-LSTM model has shown remarkable effectiveness, achieving an accuracy rate of 97.22%. The results obtained demonstrated the feasibility and effectiveness of deep learning techniques for recognizing depression-related patterns in the emotions of Twitter users. This research provides a foundation for advanced predictive analytics in mental health monitoring and highlights the potential of Bi-LSTM in improving the accuracy and efficiency of depression detection models.

Keywords

• Social Media
• Depression Classification
• Twitter
• Deep Learning
• Bi-LSTM.

Sosyal Medya Etkileşimlerinde Depresyonu Tanımlamak için Derin Öğrenme Tekniklerinin Kullanılması

Abstract

Depresyon, en yaygın zihinsel sorunlardan biridir ve intiharların önemli bir nedenidir. Sosyal medya platformlarının kullanımının artması, kullanıcıların günlük dilini kullanarak ifade ettikleri cümleler üzerinden depresyonun erken teşhisine olanak sağlamıştır. Sosyal medya platformlarının bireylerin günlük hayatlarında merkezi bir rol oynamaya devam etmesiyle, bu platformları ruh sağlığı analizi için kullanma konusunda artan bir ilgi bulunmaktadır. Bu çalışmada, Twitter (günümüzde X) üzerinden depresyon sınıflandırması yapılmıştır. Bu çalışmanın amacı Twitter'dan alınan tweetler arasında depresif özellik taşıyan tweetleri tespit etmektir. Çalışmada, Çift Yönlü Uzun Kısa Süreli Bellek (Bi-LSTM) mimarisi kullanarak depresyon tahmini için yenilikçi bir model sunulmuştur. Bu model, tweetlerdeki dil özelliklerini kullanarak depresyonun daha doğru tespiti için uygun temizleme ve ön işleme tekniklerinden faydalanmaktadır. Çalışma için, Twitter API yoluyla elde edilen özel bir veri seti oluşturulmuş ve analizler bu veri seti üzerinde gerçekleştirilmiştir. Önerilen Bi-LSTM modeli, %97.22'lik bir doğruluk oranı elde ederek dikkate değer bir etkinlik göstermiştir. Elde edilen sonuçlar, Twitter kullanıcılarının duygularındaki depresyonla ilgili örüntüleri ayırt etmek için derin öğrenme tekniklerinin kullanılabilirliği ve etkinliğini göstermiştir. Bu araştırma, ruh sağlığı izlemede ileri düzey tahmin analitikleri için bir temel oluşturmakta ve depresyon tespit modellerinin doğruluğunu ve verimliliğini artırmada Bi-LSTM'in potansiyelini vurgulamaktadır.

Keywords

• Sosyal Medya
• Depresyon Sınıflandırması
• Twitter
• Derin Öğrenme
• Bi-LSTM.

References

1. Uddin, M. Z., Dysthe, K. K., Følstad, A., & Brandtzaeg, P. B. (2022). Deep learning for prediction of depressive symptoms in a large textual dataset. Neural Computing and Applications, 34(1), 721–744. https://doi.org/10.1007/s00521-021-06426-4
2. Oquendo, M. A., Ellis, S. P., Greenwald, S., Malone, K. M., Weissman, M. M., & Mann, J. J. (2001). Ethnic and sex differences in suicide rates relative to major depression in the United States. American Journal of Psychiatry, 158(10), 1652–1658. https://doi.org/10.1176/appi.ajp.158.10.1652
3. Zafar, A., & Chitnis, S. (2020). Survey of depression detection using social networking sites via data mining. IEEE Xplore, 88-93. https://doi.org/10.1109/Confluence47617.2020.9058189
4. Martínez-Castaño, R., Pichel, J. C., & Losada, D. E. (2020). A big data platform for real time analysis of signs of depression in social media. International Journal of Environmental Research and Public Health, 17(13), 4752. https://doi.org/10.3390/ijerph17134752
5. Patel, V., Ramasundarahettige, C., Vijayakumar, L., Thakur, J., Gajalakshmi, V., Gururaj, G., Suraweera, W., & Jha, P. (2012). Suicide mortality in India: A nationally representative survey. The Lancet, 379(9834), 2343–2351. https://doi.org/10.1016/S0140-6736(12)60606-0
6. Lin, C., Hu, P., Su, H., Li, S., Mei, J., Zhou, J., & Leung, H. (2020, June 8-11). Sensemood: depression detection on social media [Conference Presentation]. International Conference on Multimedia Retrieval, Dublin Ireland. https://doi.org/10.1145/3372278.3391932
7. Conway, M. & O’Connor, D. (2016). Social media, big data, and mental health: Current advances and ethical implications. Current Opinion in Psychology, 9, 77–82. https://doi.org/10.1016/j.copsyc.2016.01.004
8. Ebert, D. D., Harrer, M., Apolinário-Hagen, J., & Baumeister, H. (2019). Digital Interventions for Mental Disorders: Key Features, Efficacy, and Potential for Artificial Intelligence Applications. In: Kim, Y. K. (ed), Frontiers in Psychiatry, (pp. 583–627). Springer.

9. Javed, A. R., Sarwar, M. U., Beg, M. O., Asim, M., Baker, T., & Tawfik, H. (2020). A collaborative healthcare framework for shared healthcare plan with ambient intelligence. Human-Centric Computing and Information Sciences, 10, 1-21. https://doi.org/10.1186/s13673-020-00245-7
10. Kale, S. S. (2015). Tracking mental disorders across Twitter users [Doctoral Dissertation, University of Georgia].
11. Coppersmith, G., Dredze, M., & Harman, C. (2014). Quantifying Mental Health Signals in Twitter. In Resnik, P., Resnik, R., and Mitchell, M.(eds), Workshop on Computational Linguistics and Clinical Psychology: From Linguistic Signal to Clinical Reality, (pp. 51-60). Association for Computational Linguistics.
12. Losada, D. E., & Crestani, F. (2016). A test collection for research on depression and language use. Experimental IR Meets Multilinguality, Multimodality, and Interaction, 28-39. https://doi.org/10.1007/978-3-319-44564-9_3
13. Ríssola, E. A., Aliannejadi, M., & Crestani, F. (2020, April 14–17). Beyond Modelling: Understanding Mental Disorders in Online Social Media. In J. M. Jose., Yilmaz, E., Magalhães, J., Castells, P., Ferro, N., Silva, M.J., Martins, F. (Eds.), Advances in Information Retrieval: 42nd European Conference on IR Research, ECIR, (pp. 296-310). Springer International Publishing. https://doi.org/10.1007/978-3-030-45439-5_20
14. Fatima, I., Abbasi, B. U. D., Khan, S., Al‐Saeed, M., Ahmad, H. F., & Mumtaz, R. (2019). Prediction of postpartum depression using machine learning techniques from social media text. Expert Systems, 36(4), e12409. https://doi.org/10.1111/exsy.12409
15. Suman, S. K., Shalu, H., Agrawal, L. A., Agrawal, A., & Kadiwala, J. (2020). A novel sentiment analysis engine for preliminary depression status estimation on social media. arXiv:2011.14280. https://doi.org/10.1111/exsy.12409
16. Zehra, W., Javed, A. R., Jalil, Z., Khan, H. U., & Gadekallu, T. R. (2021). Cross corpus multi-lingual speech emotion recognition using ensemble learning. Complex & Intelligent Systems, 7(4), 1-10. https://doi.org/10.1525/collabra.18731
17. Costello, C., Srivastava, S., Rejaie, R., & Zalewski, M. (2021). Predicting mental health from followed accounts on Twitter. Collabra: Psychology, 7(1), 18731. https://doi.org/10.1525/collabra.18731
18. Eichstaedt, J. C., Smith, R. J., Merchant, R. M., Ungar, L. H., Crutchley, P., Preoţiuc-Pietro, D., & Schwartz, H. (2018). Facebook language predicts depression in medical records. National Academy of Sciences, 115(44), 11203-11208. https://doi.org/10.1073/pnas.1802331115
19. Ahmad, H., Asghar, M. Z., Alotaibi, F. M., & Hameed, I. A. (2020). Applying deep learning technique for depression classification in social media text. Journal of Medical Imaging and Health Informatics, 10(10), 2446- 2451. https://doi.org/10.1166/jmihi.2020.3169
20. Priya, A., Garg, S., & Tigga, N. P. (2020). Predicting anxiety, depression and stress in modern life using machine learning algorithms. Procedia Computer Science, 167, 1258-1267. https://doi.org/10.1016/j.procs.2020.03.442
21. Mori, K., & Haruno, M. (2021). Differential ability of network and natural language information on social media to predict interpersonal and mental health traits. Journal of Personality, 89(2), 228-243. https://doi.org/10.1111/jopy.12578
22. Tao, X., Zhou, X., Zhang, J., & Yong, J. (2016, December 12-15). Sentiment analysis for depression detection on social networks [Conference presentation]. 12th International Conference, ADMA Australia. https://doi.org/10.1007/978-3-319-49586-6_59
23. Guntuku, S. C., Schneider, R., Pelullo, A., Young, J., Wong, V., Ungar, L., & Merchant, R. (2019). Studying expressions of loneliness in individuals using twitter: an observational study. BMJ Open, 9(11). https://doi.org/10.1136/bmjopen-2019-030355
24. Smys, D. S., & Raj, D. J. S. (2021). Analysis of deep learning techniques for early detection of depression on social media network-a comparative study. Journal of Trends in Computer Science and Smart Technology, 3(1), 24-39. https://doi.org/10.36548/jtcsst.2021.1.003
25. Hiraga, M. (2017). Predicting depression for japanese blog text. Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics- Student Research Workshop, 107–113 https://doi.org/10.18653/v1/P17-3018 P
26. Wu, J., Ma, J., Wang, Y., & Wang, J. (2021). Understanding and predicting the burst of burnout via social media. Proceedings of the ACM on Human-Computer Interaction, 4(CSCW3), 1-27. https://doi.org/10.1145/3434174
27. Orabi, A. H., Buddhitha, P., Orabi, M. H., & Inkpen, D. (2018). Deep learning for depression detection of twitter users. Proceedings of the Fifth Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, 88-97. https://doi.org/10.18653/v1/W18-0609
28. Shah, F. M., Ahmed, F., Joy, S. K. S., Ahmed, S., Sadek, S., Shil, R., & Kabir, M. H. (2020). Early depression detection from social network using deep learning techniques, IEEE Region 10 Symposium (TENSYMP), 823-826. https://doi.org/10.1109/TENSYMP50017.2020.9231008
29. Shen, G., Jia, J., Nie, L., Feng, F., Zhang, C., Hu, T., & Zhu, W. (2017). Depression detection via harvesting social media: A multimodal dictionary learning solution. Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17), 3838-3844. https://doi.org/10.24963/ijcai.2017/536
30. De Choudhury, M., Gamon, M., Counts, S., & Horvitz, E. (2013). Predicting depression via social media. Proceedings of the International AAAI Conference on Web and Social Media, 7(1), 128-137. https://doi.org/10.1609/icwsm.v7i1.14432
31. Aldarwish, M. M., & Ahmad, H. F. (2017). Predicting depression levels using social media posts. 2017 IEEE 13th International Symposium on Autonomous Decentralized System (ISADS). 277-280. https://doi.org/10.1109/ISADS.2017.41
32. Guntuku, S. C., Yaden, D. B., Kern, M. L., Ungar, L. H., & Eichstaedt, J. C. (2017). Detecting depression and mental illness on social media: an integrative review. Current Opinion in Behavioral Sciences, 18, 43-49. https://doi.org/10.1016/j.cobeha.2017.07.005
33. Ahmad, H., Asghar, M. Z., Alotaibi, F. M., & Hameed, I. A. (2020). Applying deep learning technique for depression classification in social media text. Journal of Medical Imaging and Health Informatics, 10(10), 2446-2451. https://doi.org/10.1166/jmihi.2020.3169
34. Figuerêdo, José Solenir L., Ana Lúcia L. M. Maia, & Rodrigo Tripodi Calumby. (2022). Early depression detection in social media based on deep learning and underlying emotions. Online Social Networks and Media, 31, 100225. https://doi.org/10.1016/j.osnem.2022.100225
35. Marriwala, Nikhil, & Deepti Chaudhary. (2023). A hybrid model for depression detection using deep learning. Measurement: Sensors, 25, 100587. https://doi.org/10.1016/j.measen.2022.100587
36. Yapıcı, M. M. (2022). Lojistik Regresyon Modeli. In Savaş, S. & Buyrukoğlu, S., (Eds). Teori ve Uygulamada Makine Öğrenmesi (pp 37-67). Ankara: Nobel Akademik Yayıncılık Eğitim Danışmanlık TİC. LTD. ŞTİ.
37. Karakış, R. (2022). Destek Vektör Makinesi In Savaş, S. & Buyrukoğlu, S., (Eds). Teori ve Uygulamada Makine Öğrenmesi (pp 93-118). Ankara: Nobel Akademik Yayıncılık Eğitim Danışmanlık TİC. LTD. ŞTİ.
38. Schwartz, H. A., Eichstaedt, J., Kern, M., Park, G., Sap, M., Stillwell, D., & Ungar, L. (2014). Towards assessing changes in degree of depression through facebook. In Resnik, P., Resnik, R., and Mitchell, M.(eds), Proceedings of the workshop on computational linguistics and clinical psychology: from linguistic signal to clinical reality. (pp 118–125) Association for Computational Linguistics. https://doi.org/10.3115/v1/W14-3214