LDA-SVM-based topic modeling for healthcare document classification and medical decision support in low-resource settings
DOI:
https://doi.org/10.20372/star.V15.i3.08Keywords:
EHRs, Topic Modeling, LDA, SVM, Classification, Medical Decision SupportAbstract
There is great potential and many challenges in the increased number of texts related to the field of healthcare, especially in low-resource languages like Afaan Oromo. The proper classification and retrieval of the text documents is very important in making a medical decision. This paper provides a hybrid approach using Latent Dirichlet Allocation (LDA) for interpretation of topics and Support Vector Machine for proper classification of the documents. The use of the hybrid approach on the healthcare documents will enable us to have access to relevant information in linguistically poor areas. We used the EHR dataset in this study and applied different pre-processing methods to it. A feature extraction technique was utilized. In this case, the developed model LDA-SVM has shown good performance, achieving an accuracy of 93.67% and clear interpretation of topics. This model has been successful in classifying EHRs, especially in dealing with unlabeled datasets. But in this research, only a monolingual model was developed for the Afaan Oromo language. Expanding this work to support multilingual, including code-mixed and cross-lingual data, remains an important direction for future research.
Downloads
Metrics
References
Blei, D. M., Ng, A. Y., & Jordan, M. I. (2003). Latent Dirichlet Allocation. The Journal of Machine Learning Research 3, 993–1022.
https://dl.acm.org/doi/10.5555/944919.944937
Cahyo, P. W., Aesyi, U. S., & Santosa, B. D. (2024). Topic Sentiment Using Logistic Regression and Latent Dirichlet Allocation as a Customer Satisfaction Analysis Model. Jurnal Infotel, 16(1), 1–16.
https://doi.org/10.20895/infotel.v16i1.1081
Chauhan, U., & Shah, A. (2021). Topic Modeling Using Latent Dirichlet allocation. ACM Computing Surveys, 54(7), 1–35.
https://doi.org/10.1145/3462478
Dinsa, E. F., Das, M., & Abebe, T. U. (2024a). A topic modeling approach for analyzing and categorizing electronic healthcare documents in Afaan Oromo without label information. Scientific Reports, 14(1), 1–14.
https://doi.org/10.1038/s41598-024-83743-3
Dinsa, E. F., Das, M., & Abebe, T. U. (2024b). AI-based disease category prediction model using symptoms from low-resource Ethiopian language: Afaan Oromo text. Scientific Reports, 14(1), 11233.
https://doi.org/10.1038/s41598-024-62278-7
Dinsa, E. F., Das, M., & Abebe, T. U. (2026). Advances in electronic health records enabled by artificial intelligence and natural language processing: a review of recent developments, limitations and future applications. Discover Applied Sciences, 8(3), 250.
https://doi.org/10.1007/s42452-026-08318-5
Dinsa, E. F., Das, M., Abebe, T. U., & Ramaswamy, K. (2024). Automatic categorization of medical documents in Afaan Oromo using ensemble machine learning techniques. Discover Applied Sciences, 6(11). 575.
https://doi.org/10.1007/s42452-024-06307-0
Dinsa, E. F., & P, R. B. (2019). Application of data mining Classification Algorithms for AFAAN Oromo Media text news Categorization. International Journal of Computer Trends and Technology, 67(7), 73–79. https://doi.org/10.14445/22312803/ijctt-v672
Farkhod, A., Abdusalomov, A., Makhmudov, F., & Cho, Y. I. (2021). LDA-Based Topic Modeling Sentiment Analysis using Topic/Document/Sentence (TDS) model. Applied Sciences, 11(23), 11091.
https://doi.org/10.3390/app112311091
Fikadu, E. (2020). Comparative study of the Performance of Machine Learning Text Classifiers Applied to Afaan Oromo Text. International Journal of Scientific Research in Computer Science Engineering and Information Technology, 77–83. https://doi.org/10.32628/cseit20645
Fikadu, E., Das, M., Urgessa, T., & Ramaswamy, K. (2025). A supervised learning approach for recommending medical specialists in the healthcare sector for the Afaan Oromo context. Discover Computing, 28(1), 46.
https://doi.org/10.1007/s10791-025-09556-8
Fikadu, E., Urgessa, T., & Das, M. (2025). MLP-SVM: a hybrid approach for improving the performance of the classification model for health-related documents from social media using multi-layer perceptron and support vector machine. Discover Applied Sciences, 7(4), 320. https://doi.org/10.1007/s42452-025-06851-3
Hu, Y., Boyd-Graber, J., Satinoff, B., & Smith, A. (2014). Interactive topic modeling. Machine Learning, 95(3), 423–469.
https://doi.org/10.1007/s10994-013-5413-0
Lu, Y. (2013). Automatic topic identification of health-related messages in online health community using text classification. SpringerPlus, 2(1), 1–7.
https://doi.org/10.1186/2193-1801-2-309
Muthusami, R., Mani Kandan, N., Saritha, K., Narenthiran, B., Nagaprasad, N., & Ramaswamy, K. (2024). Investigating topic modeling techniques through evaluation of topics discovered in short texts data across diverse domains. Scientific Reports, 14(1), 12003.
https://doi.org/10.1038/s41598-024-61738-4
Negash Tesso, D., Fikadu Dinsa, E., & Fikadu Kenani, H. (2023). Signed Language Translation into Afaan Oromo Text Using Deep-Learning Approach. American Journal of Artificial Intelligence, 7(2), 40–51. https://doi.org/10.11648/j.ajai.20230702.12
Rogers, N., & Longo, L. (2017). A comparison on the classification of short-text documents using Latent Dirichlet Allocation and Formal Concept Analysis. CEUR Workshop Proceedings, 2086(December), 50–62.
https://ceur-ws.org/Vol-2086/AICS2017_pa per_2 7.pdf
Shakeel, K., Tahir, G. R., Tehseen, I., & Ali, M. (2018). A framework of Urdu topic modeling using latent dirichlet allocation (LDA). 2018 IEEE 8th Annual Computing and Communication Workshop and Conference, CCWC 2018, 2018-Janua(June 2020), 117–123. https://doi.org/10.1109/CCWC.2018.8301655
Stewart, J., Lu, J., Goudie, A., Arendts, G., Meka, S. A., Freeman, S., Walker, K., Sprivulis, P., Sanfilippo, F., Bennamoun, M., & Dwivedi, G. (2023). Applications of natural language processing at emergency department triage: A narrative review. PLoS ONE, 18(12), 1–25. https://doi.org/10.1371/journal.pone.0279953
Sumathi, M., & Raja, S. P. (2023). Machine learning algorithm-based spam detection in social networks. Social Network Analysis and Mining, 13(1), 1–13.
https://doi.org/10.1007/s13278-023-01108-6
Thielmann, A., Weisser, C., Krenz, A., & Säfken, B. (2023). Unsupervised document classification integrating web scraping, one-class SVM and LDA topic modelling. Journal of Applied Statistics, 50(3), 574–591. https://doi.org/10.1080/02664763.2021.19190
Wu, M., Du, X., Gu, R., & Wei, J. (2021). Artificial Intelligence for Clinical Decision Support in Sepsis. Frontiers in Medicine, 8(5), 1–9. https://doi.org/10.3389/fmed.2021.665464
Yang, S. and Zhang, H. (2018). Text Mining of Twitter Data Using a Latent Dirichlet Allocation Topic Model and Sentiment Analysis. International Journal of Computer and Information Engineering, 12(7), 525–529.
Downloads
Published
How to Cite
License
Copyright (c) 2026 Science, Technology and Arts Research Journal
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
STAR © Copyright; All rights reserved
