{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,6]],"date-time":"2026-07-06T00:16:18Z","timestamp":1783296978434,"version":"3.54.6"},"reference-count":25,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T00:00:00Z","timestamp":1777852800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,7,6]],"date-time":"2026-07-06T00:00:00Z","timestamp":1783296000000},"content-version":"vor","delay-in-days":63,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Sci Rep"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>The case of mental health disorders has been a main topic in the clinical and psychological field. The advancement of computing studies, especially in Natural Language Processing (NLP)\u2014a subset of Machine Learning, created a system of detection that can detect the mental health state of a person in early stage to prevent the eventuality of the worst case. This is crucial since there has been a lot of case of mental health disorder\u2014such as depression and suicide, remains undetected and untreated\u2013especially when the internet usage is more prevalent than ever even among the most vulnerable users, which are the preadolescent users. This study explores the models that can accurately predict mental health disorder with the provided six labels the model can predict. The labels are anxiety, depression, personality disorder, stress, bipolar, and normal. The dataset is gathered from a Kaggle repository which is then processed and refined further for the training process. From multiple evaluations across diverse amount of texts from different users, our Bi-LSTM-XGBoost model outperforms the other models with an accuracy of 0.9035 and 0.4320 loss, while other models fall short within 50\u201384% accuracy. Further improvement can be made with our model, whether from improving the model\u2019s parameters further or by improving the quantity and quality of the dataset gathered.<\/jats:p>","DOI":"10.1038\/s41598-026-47015-6","type":"journal-article","created":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T14:06:12Z","timestamp":1777903572000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Early detection of mental health on social media using a hybrid Bi-LSTM\u2013XGBoost model: a comparative study"],"prefix":"10.1038","volume":"16","author":[{"given":"Heru Syah","family":"Putra","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Muhammad Ary","family":"Murti","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Astri","family":"Novianty","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nur Hafieza","family":"Ismail","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,5,4]]},"reference":[{"key":"47015_CR1","doi-asserted-by":"publisher","first-page":"458","DOI":"10.1093\/jurban\/78.3.458","volume":"78","author":"I Kawachi","year":"2001","unstructured":"Kawachi, I. & Berkman, L. F. Social ties and mental health. J. Urban Health 78, 458\u2013467. https:\/\/doi.org\/10.1093\/jurban\/78.3.458 (2001).","journal-title":"J. Urban Health"},{"key":"47015_CR2","doi-asserted-by":"publisher","unstructured":"Frank, R.\u00a0G. & McGuire, T.\u00a0G. Chapter 16 economics and mental health. vol.\u00a01 of Handbook of Health Economics, 893\u2013954. https:\/\/doi.org\/10.1016\/S1574-0064(00)80029-3 (Elsevier, 2000).","DOI":"10.1016\/S1574-0064(00)80029-3"},{"key":"47015_CR3","doi-asserted-by":"publisher","DOI":"10.1016\/j.cobeha.2022.101145","volume":"45","author":"I Cataldo","year":"2022","unstructured":"Cataldo, I., Billieux, J., Esposito, G. & Corazza, O. Assessing problematic use of social media: where do we stand and what can be improved?. Curr. Opin. Behav. Sci. 45, 101145. https:\/\/doi.org\/10.1016\/j.cobeha.2022.101145 (2022).","journal-title":"Curr. Opin. Behav. Sci."},{"key":"47015_CR4","doi-asserted-by":"publisher","DOI":"10.1016\/j.abrep.2020.100281","volume":"12","author":"S Casale","year":"2020","unstructured":"Casale, S. Problematic social media use: Conceptualization, assessment and trends in scientific literature. Addict. Behav. Rep. 12, 100281. https:\/\/doi.org\/10.1016\/j.abrep.2020.100281 (2020).","journal-title":"Addict. Behav. Rep."},{"key":"47015_CR5","doi-asserted-by":"publisher","unstructured":"Khan, A., & Ali, R. Unraveling minds in the digital era: a review on mapping mental health disorders through machine learning techniques using online social media. Soc. Netw. Anal. Min. 14, https:\/\/doi.org\/10.1007\/s13278-024-01205-0 (2024).","DOI":"10.1007\/s13278-024-01205-0"},{"key":"47015_CR6","doi-asserted-by":"publisher","first-page":"129","DOI":"10.1186\/s40337-024-01096-8","volume":"12","author":"J Portingale","year":"2024","unstructured":"Portingale, J., Girardin, S., Liu, S., Fuller-Tyszkiewicz, M. & Krug, I. Daily bi-directional effects of women\u2019s social media-based appearance comparisons, body satisfaction, and disordered eating urges. J. Eat. Disord. 12, 129. https:\/\/doi.org\/10.1186\/s40337-024-01096-8 (2024).","journal-title":"J. Eat. Disord."},{"key":"47015_CR7","doi-asserted-by":"publisher","unstructured":"Oprea, S.-V. & B\u00e2ra, A. Assessing the dual impact of the social media platforms on psychological well-being: A multiple-option descriptive-predictive framework. Comput. Econ. 66, https:\/\/doi.org\/10.1007\/s10614-024-10717-y (2025).","DOI":"10.1007\/s10614-024-10717-y"},{"key":"47015_CR8","doi-asserted-by":"crossref","unstructured":"Chen, H., McKeever, S. & Delany, S.\u00a0J. A comparison of classical versus deep learning techniques for abusive content detection on social media sites. In Staab, S., Koltsova, O. & Ignatov, D.\u00a0I. (eds.) Social Informatics, 117\u2013133 (Springer International Publishing, Cham, 2018).","DOI":"10.1007\/978-3-030-01129-1_8"},{"key":"47015_CR9","doi-asserted-by":"publisher","first-page":"196","DOI":"10.1007\/s13278-024-01360-4","volume":"14","author":"M Kerasiotis","year":"2024","unstructured":"Kerasiotis, M., Ilias, L. & Askounis, D. Depression detection in social media posts using transformer-based models and auxiliary features. Soc. Netw. Anal. Min. 14, 196. https:\/\/doi.org\/10.1007\/s13278-024-01360-4 (2024).","journal-title":"Soc. Netw. Anal. Min."},{"key":"47015_CR10","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2021.104499","volume":"135","author":"R Chiong","year":"2021","unstructured":"Chiong, R., Budhi, G. S., Dhakal, S. & Chiong, F. A textual-based featuring approach for depression detection using machine learning classifiers and social media texts. Comput. Biol. Med. 135, 104499. https:\/\/doi.org\/10.1016\/j.compbiomed.2021.104499 (2021).","journal-title":"Comput. Biol. Med."},{"key":"47015_CR11","doi-asserted-by":"publisher","first-page":"360","DOI":"10.1016\/j.jad.2020.08.091","volume":"286","author":"J Kim","year":"2021","unstructured":"Kim, J. et al. A systematic review of the validity of screening depression through facebook, twitter, instagram, and snapchat. J. Affect. Disord. 286, 360\u2013369. https:\/\/doi.org\/10.1016\/j.jad.2020.08.091 (2021).","journal-title":"J. Affect. Disord."},{"key":"47015_CR12","doi-asserted-by":"publisher","first-page":"6","DOI":"10.1007\/s13755-022-00197-5","volume":"11","author":"Z Li","year":"2023","unstructured":"Li, Z. et al. Mha: a multimodal hierarchical attention model for depression detection in social media. Health Inf. Sci. Syst. 11, 6. https:\/\/doi.org\/10.1007\/s13755-022-00197-5 (2023).","journal-title":"Health Inf. Sci. Syst."},{"key":"47015_CR13","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1007\/s13755-024-00303-9","volume":"12","author":"E Bao","year":"2024","unstructured":"Bao, E., P\u00e9rez, A. & Parapar, J. Explainable depression symptom detection in social media. Health Inf. Sci. Syst. 12, 47. https:\/\/doi.org\/10.1007\/s13755-024-00303-9 (2024).","journal-title":"Health Inf. Sci. Syst."},{"key":"47015_CR14","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.neubiorev.2019.06.032","volume":"104","author":"M Brand","year":"2019","unstructured":"Brand, M. et al. The interaction of person-affect-cognition-execution (i-pace) model for addictive behaviors: Update, generalization to addictive behaviors beyond internet-use disorders, and specification of the process character of addictive behaviors. Neurosci. Biobehav. Rev. 104, 1\u201310. https:\/\/doi.org\/10.1016\/j.neubiorev.2019.06.032 (2019).","journal-title":"Neurosci. Biobehav. Rev."},{"key":"47015_CR15","doi-asserted-by":"publisher","first-page":"459","DOI":"10.1007\/s40429-021-00400-9","volume":"8","author":"A Brandtner","year":"2021","unstructured":"Brandtner, A., Antons, S., Cornil, A. & Brand, M. Integrating desire thinking into the i-pace model: a special focus on internet-use disorders. Curr. Addict. Rep. 8, 459\u2013468. https:\/\/doi.org\/10.1007\/s40429-021-00400-9 (2021).","journal-title":"Curr. Addict. Rep."},{"key":"47015_CR16","doi-asserted-by":"publisher","DOI":"10.1016\/j.addbeh.2024.108044","volume":"155","author":"D Wei","year":"2024","unstructured":"Wei, D., Chan, L.-S., Du, N., Hu, X. & Huang, Y.-T. Gratification and its associations with problematic internet use: A systematic review and meta-analysis using use and gratification theory. Addict. Behav. 155, 108044. https:\/\/doi.org\/10.1016\/j.addbeh.2024.108044 (2024).","journal-title":"Addict. Behav."},{"key":"47015_CR17","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2024.123834","volume":"249","author":"J Philip Thekkekara","year":"2024","unstructured":"Philip Thekkekara, J., Yongchareon, S. & Liesaputra, V. An attention-based cnn-bilstm model for depression detection on social media text. Expert Syst. Appl. 249, 123834. https:\/\/doi.org\/10.1016\/j.eswa.2024.123834 (2024).","journal-title":"Expert Syst. Appl."},{"key":"47015_CR18","doi-asserted-by":"publisher","unstructured":"Aggarwal, R. & Goyal, A. Anxiety and depression detection using machine learning. In 2022 International Conference on Machine Learning, Big Data, Cloud and Parallel Computing (COM-IT-CON), vol.\u00a01, 141\u2013149, https:\/\/doi.org\/10.1109\/COM-IT-CON54601.2022.9850532 (2022).","DOI":"10.1109\/COM-IT-CON54601.2022.9850532"},{"key":"47015_CR19","doi-asserted-by":"publisher","first-page":"134113","DOI":"10.1109\/ACCESS.2022.3231884","volume":"10","author":"A-H Jo","year":"2022","unstructured":"Jo, A.-H. & Kwak, K.-C. Diagnosis of depression based on four-stream model of bi-lstm and cnn from audio and text information. IEEE Access 10, 134113\u2013134135. https:\/\/doi.org\/10.1109\/ACCESS.2022.3231884 (2022).","journal-title":"IEEE Access"},{"key":"47015_CR20","doi-asserted-by":"publisher","first-page":"85428","DOI":"10.1109\/ACCESS.2025.3568340","volume":"13","author":"MM Hossain","year":"2025","unstructured":"Hossain, M. M. et al. Revolutionizing mental health sentiment analysis with bert-fuse: A hybrid deep learning model. IEEE Access 13, 85428\u201385446. https:\/\/doi.org\/10.1109\/ACCESS.2025.3568340 (2025).","journal-title":"IEEE Access"},{"key":"47015_CR21","doi-asserted-by":"publisher","unstructured":"Odja, K.\u00a0D., Widiarta, J., Purwanto, E.\u00a0S. & Ario, M.\u00a0K. Mental illness detection using sentiment analysis in social media. Proc. Comput. Sci. 245, 971\u2013978, https:\/\/doi.org\/10.1016\/j.procs.2024.10.325 (2024). 9th International Conference on Computer Science and Computational Intelligence 2024 (ICCSCI 2024).","DOI":"10.1016\/j.procs.2024.10.325"},{"key":"47015_CR22","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2024.104586","volume":"150","author":"M Garg","year":"2024","unstructured":"Garg, M., Liu, X., Sathvik, M., Raza, S. & Sohn, S. Multiwd: Multi-label wellness dimensions in social media posts. J. Biomed. Inform. 150, 104586. https:\/\/doi.org\/10.1016\/j.jbi.2024.104586 (2024).","journal-title":"J. Biomed. Inform."},{"key":"47015_CR23","doi-asserted-by":"publisher","unstructured":"Rai, S. et al. Key language markers of depression on social media depend on race. Proc. Natl. Acad. Sci. 121, e2319837121. https:\/\/doi.org\/10.1073\/pnas.2319837121 (2024).","DOI":"10.1073\/pnas.2319837121"},{"key":"47015_CR24","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","volume":"9","author":"S Hochreiter","year":"1997","unstructured":"Hochreiter, S. & Schmidhuber, J. Long short-term memory. Neural Comput. 9, 1735\u20131780. https:\/\/doi.org\/10.1162\/neco.1997.9.8.1735 (1997).","journal-title":"Neural Comput."},{"key":"47015_CR25","doi-asserted-by":"publisher","unstructured":"Graves, A. & Schmidhuber, J. Framewise phoneme classification with bidirectional lstm networks. In Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005., vol.\u00a04, 2047\u20132052 vol. 4, https:\/\/doi.org\/10.1109\/IJCNN.2005.1556215 (2005).","DOI":"10.1109\/IJCNN.2005.1556215"}],"container-title":["Scientific Reports"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-47015-6","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-47015-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-47015-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,6]],"date-time":"2026-07-06T00:10:24Z","timestamp":1783296624000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-47015-6"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,4]]},"references-count":25,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["47015"],"URL":"https:\/\/doi.org\/10.1038\/s41598-026-47015-6","relation":{},"ISSN":["2045-2322"],"issn-type":[{"value":"2045-2322","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,4]]},"assertion":[{"value":"10 December 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"29 March 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"4 May 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}},{"value":"As the dataset was sourced from a publicly available Kaggle repository, no direct web scraping or human-subject data collection was performed by the authors. The dataset does not contain any personally identifiable information about the original posters. Furthermore, the proposed model is designed strictly for early screening support and risk prioritization, and must not be used as a definitive clinical diagnosis tool.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics and privacy statement"}}],"article-number":"20582"}}