Predicción temprana de brotes de dengue en Bucaramanga mediante un modelo de Machine Learning con integración de variables climáticas, geográficas y sociodemográficas
| dc.contributor.advisor | Álvarez Pizarro, Yuli Andrea | |
| dc.contributor.author | Araque Villamizar, Reinaldo | |
| dc.contributor.author | Paipa Bautista, Diego Alberto | |
| dc.contributor.corporatename | Universidad Santo Tomás | |
| dc.date.accessioned | 2026-08-28T19:40:23Z | |
| dc.date.available | 2026-08-28T19:40:23Z | |
| dc.date.issued | 2026-08-28 | |
| dc.description | Problema. El dengue es una arbovirosis endémica en Bucaramanga que entre 2015 y 2022 acumuló 10.259 casos notificados al SIVIGILA, con una tasa de hospitalización del 26,6 % y picos epidémicos recurrentes; sin embargo, el municipio no dispone de herramientas que anticipen sus brotes. Objetivo. Desarrollar un modelo de machine learning para la predicción temprana de brotes de dengue en Bucaramanga, integrando variables epidemiológicas, climáticas, geográficas y sociodemográficas. Método. Investigación cuantitativa, aplicada y longitudinal que, siguiendo el proceso CRISP-DM, integró datos del SIVIGILA, registros climáticos del IDEAM, índices de vegetación (NDVI) de Sentinel-2 e indicadores del DANE, agregados por semana epidemiológica y comuna. Se compararon cinco enfoques —un modelo naïve estacional y ARIMA como líneas base frente a Random Forest, XGBoost y LSTM— mediante una partición temporal estricta y horizontes de predicción de una a cuatro semanas, evaluados con las métricas MAE, RMSE, MAPE, MASE, F1 y AUC, la prueba de Diebold-Mariano y la interpretabilidad SHAP. Resultados. Los modelos de machine learning redujeron el error de predicción a menos de dos tercios del pronóstico ingenuo y discriminaron las semanas de brote de forma estadísticamente significativa (p < 0,001). La temperatura, con un rezago de siete semanas, fue el predictor ambiental más relevante, y el riesgo se concentró en las comunas San Francisco, Centro y Oriental (32,4 % de los casos). Conclusiones. Es factible anticipar los brotes con datos públicos locales; el prototipo DengueAlert-BGA constituye una herramienta interpretable y replicable para fortalecer la vigilancia y la respuesta sanitaria del municipio. | |
| dc.description.abstract | Problem. Dengue is an endemic arboviral disease in Bucaramanga that accumulated 10,259 cases reported to the National Public Health Surveillance System (SIVIGILA) between 2015 and 2022, with a hospitalization rate of 26.6% and recurrent epidemic peaks. However, the municipality lacks tools capable of anticipating outbreaks. Objective. To develop a machine learning model for the early prediction of dengue outbreaks in Bucaramanga by integrating epidemiological, climatic, geographic, and sociodemographic variables. Methods. A quantitative, applied, and longitudinal study was conducted following the CRISP-DM methodology. Data from SIVIGILA, climate records from IDEAM, Sentinel-2 vegetation indices (NDVI), and sociodemographic indicators from DANE were integrated and aggregated by epidemiological week and administrative district. Five approaches were compared: a seasonal naïve model and ARIMA as baseline models, versus Random Forest, XGBoost, and LSTM. Models were evaluated using a strict temporal split and prediction horizons ranging from one to four weeks. Performance was assessed using MAE, RMSE, MAPE, MASE, F1-score, and AUC metrics, complemented by the Diebold–Mariano test and SHAP-based interpretability analysis. Results. Machine learning models reduced prediction error to less than two-thirds of the naïve forecast and significantly discriminated outbreak weeks (p < 0.001). Temperature, with a seven-week lag, was identified as the most relevant environmental predictor. Outbreak risk was concentrated in the San Francisco, Centro, and Oriental districts, which accounted for 32.4% of reported cases. Conclusions. Dengue outbreaks can be anticipated using publicly available local data. The DengueAlert-BGA prototype represents an interpretable and replicable tool to strengthen epidemiological surveillance and public health response in the municipality. | |
| dc.description.degreelevel | Maestría | spa |
| dc.description.degreename | Magíster en Análisis de Datos y Sistemas Inteligentes | spa |
| dc.description.domain | https://www.ustabuca.edu.co/ | |
| dc.format.mimetype | application/pdf | |
| dc.identifier.citation | Araque Villamizar, R., y Paipa Bautista, D. A. (2026). Predicción temprana de brotes de dengue en Bucaramanga mediante un modelo de machine learning con integración de variables climáticas, geográficas y sociodemográficas [Tesis de maestría]. Universidad Santo Tomás, Bucaramanga, Colombia. | |
| dc.identifier.instname | instname:Universidad Santo Tomás | spa |
| dc.identifier.reponame | reponame:Repositorio Institucional Universidad Santo Tomás | spa |
| dc.identifier.repourl | repourl:https://repository.usta.edu.co | spa |
| dc.identifier.uri | http://hdl.handle.net/11634/74014 | |
| dc.language.iso | spa | spa |
| dc.publisher | Universidad Santo Tomás | spa |
| dc.publisher.branch | CRAI-USTA Bucaramanga | |
| dc.publisher.faculty | Facultad de Ingeniería Mecatrónica | spa |
| dc.publisher.program | Maestría Análisis de Datos y Sistemas Inteligentes | spa |
| dc.relation.references | Alves, M. B., Erbisti, R. S., Nobre, A. A., Simões, T. C., Tavares, A. M., Melo, M. C., Pedreira, R. M., Araújo, J. P. M., Carvalho, M. S., y Honório, N. A. (2025). ARBOALVO: A Bayesian spatiotemporal learning and predictive model for dengue cases in the endemic Northeast city of Natal, Rio Grande do Norte, Brazil. PLOS Neglected Tropical Diseases, 19(4), e0012984. https://doi.org/10.1371/journal.pntd.0012984 | |
| dc.relation.references | Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324 | |
| dc.relation.references | Chen, T., y Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794. https://doi.org/10.1145/2939672.2939785 | |
| dc.relation.references | Chen, X., y Moraga, P. (2025). Forecasting dengue across Brazil with LSTM neural networks and SHAP-driven lagged climate and spatial effects. BMC Public Health, 25(1), Artículo 973. https://doi.org/10.1186/s12889-025-22106-7 | |
| dc.relation.references | Chretien, J. P., Anyamba, A., Small, J., Britch, S., Sanchez, J. L., Halbach, A. C., Tucker, C., y Linthicum, K. J. (2015a). Global climate anomalies and potential infectious disease risks: 2014–2015. PLOS Currents Outbreaks, 1. https://doi.org/10.1371/currents.outbreaks.95fbc4a8fb4695e049baabfc2fc8289f | |
| dc.relation.references | Chretien, J. P., Swedlow, D., Eckstrand, I., George, D., Shaman, J., y Linthicum, K. J. (2015b). Advancing epidemic prediction and forecasting: A new US government initiative. Online Journal of Public Health Informatics, 7(1), e13. | |
| dc.relation.references | Codeço, C. T., Coelho, F. C., Cruz, O. G., Oliveira, S. C., Castro, T. S., y Bastos, L. S. (2018). Infodengue: A nowcasting system for the surveillance of dengue fever transmission. Revista de Saúde Pública, 52, 046193. | |
| dc.relation.references | Colón-González, F. J., Gibb, R., Khan, K., Edson, C. O., Kucharski, A. J., Simons, D., y Brady, O. J. (2021). Projecting the future incidence and burden of dengue in Southeast Asia. Nature Communications, 12(1), 6335. | |
| dc.relation.references | Diebold, F. X., y Mariano, R. S. (1995). Comparing predictive accuracy. Journal of Business & Economic Statistics, 13(3), 253–263. https://doi.org/10.1080/07350015.1995.10524599 | |
| dc.relation.references | Hochreiter, S., y Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735 | |
| dc.relation.references | Instituto Nacional de Salud [INS]. (2024a). Boletín epidemiológico semanal: dengue en Colombia. Ministerio de Salud y Protección Social de Colombia. https://www.ins.gov.co | |
| dc.relation.references | Instituto Nacional de Salud [INS]. (2024b). Protocolo de vigilancia en salud pública: dengue. Ministerio de Salud y Protección Social de Colombia. https://www.ins.gov.co | |
| dc.relation.references | Johansson, M. A., Reich, N. G., Hota, A., Brownstein, J. S., y Santillana, M. (2016). Evaluating the performance of infectious disease forecasts: A comparison of climate-driven and seasonal models for dengue in Thailand. PLOS Neglected Tropical Diseases, 10(1), e0004322. https://doi.org/10.1371/journal.pntd.0004322 | |
| dc.relation.references | Lober, L., Roster, K. O., y Rodrigues, F. A. (2024). Forecasting infectious diseases in Brazilian cities: Integrating socio-economic and geographic data through machine learning. Chaos, Solitons & Fractals, 185, 115417. https://doi.org/10.1016/j.chaos.2024.115417 | |
| dc.relation.references | Lundberg, S. M., y Lee, S. I. (2017). A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 30, 4765–4774. | |
| dc.relation.references | Luz, P. M., Mendes, B. V., Codeço, C. T., Struchiner, C. J., y Galvani, A. P. (2008). Time series analysis of dengue incidence in Rio de Janeiro, Brazil. The American Journal of Tropical Medicine and Hygiene, 79(6), 933–939. | |
| dc.relation.references | Majeed, M., Aslam, M. F., Anwar, S., y Naveed, A. (2023). A deep learning approach for dengue fever prediction in Malaysia using LSTM with spatial attention. Computers in Biology and Medicine, 158, 106920. | |
| dc.relation.references | Sang, S., Gu, S., Bi, P., Yang, W., Yang, Z., Xu, L., y Liu, Q. (2015a). Predicting unprecedented dengue outbreak using imported cases and climatic factors in Guangzhou, 2014. BMC Infectious Diseases, 15(1), 166. https://doi.org/10.1186/s12879-015-0895-7 | |
| dc.relation.references | Sang, S., Yin, W., Bi, P., Zhang, H., Wang, C., Liu, X., Yang, Z., y Liu, Q. (2015b). Predicting dengue outbreaks in Guangzhou, China, using BP neural network with weather factors. Epidemiology and Infection, 143(8), 1731–1740. https://doi.org/10.1017/S095026881400236X | |
| dc.relation.references | Sebastianelli, A., Spiller, D., Carmo, R., Wheeler, J., Nowakowski, A., Jacobson, L. V., Kim, D., Barlevi, H., Cordero, Z. E. R., Colón-González, F. J., Lowe, R., Ullo, S. L., y Schneider, R. (2024). A reproducible ensemble machine learning approach to forecast dengue outbreaks. Scientific Reports, 14, Artículo 3807. https://doi.org/10.1038/s41598-024-52796-9 | |
| dc.relation.references | Zhao, N., Charland, K., Carabali, M., Nikiema, A., Maheu-Giroux, M., Rees, E., Yuan, M., Garcia, E., Zinszer, K., y GEMSTONE Consortium. (2020). Machine learning and dengue forecasting: Comparing random forests and artificial neural networks for predicting dengue burden at national and sub-national scales in Colombia. PLOS Neglected Tropical Diseases, 14(9), e0008056. https://doi.org/10.1371/journal.pntd.0008056 | |
| dc.rights.accessrights | info:eu-repo/semantics/openAccess | spa |
| dc.rights.coar | http://purl.org/coar/access_right/c_abf2 | |
| dc.rights.local | Abierto (Texto Completo) | spa |
| dc.subject.keyword | dengue | |
| dc.subject.keyword | outbreak prediction | |
| dc.subject.keyword | machine learning | |
| dc.subject.keyword | epidemiological surveillance | |
| dc.subject.keyword | early warning system | |
| dc.subject.proposal | dengue | |
| dc.subject.proposal | predicción de brotes | |
| dc.subject.proposal | machine learning | |
| dc.subject.proposal | vigilancia epidemiológica | |
| dc.subject.proposal | alerta temprana | |
| dc.title | Predicción temprana de brotes de dengue en Bucaramanga mediante un modelo de Machine Learning con integración de variables climáticas, geográficas y sociodemográficas | |
| dc.type | master thesis | |
| dc.type.category | Formación de Recurso Humano para la Ctel: Trabajo de grado de Maestría | |
| dc.type.coar | http://purl.org/coar/resource_type/c_bdcc | spa |
| dc.type.coarversion | http://purl.org/coar/version/c_ab4af688f83e57aa | spa |
| dc.type.drive | info:eu-repo/semantics/masterThesis | spa |
| dc.type.local | Tesis de maestría | spa |
| dc.type.version | info:eu-repo/semantics/acceptedVersion | spa |
Archivos
Bloque original
1 - 3 de 3
Cargando...
- Nombre:
- 2026PaipaDiego.pdf
- Tamaño:
- 2.12 MB
- Formato:
- Adobe Portable Document Format
- Descripción:
- Trabajo de grado
Cargando...
- Nombre:
- 2026PaipaDiego2.pdf
- Tamaño:
- 2.24 MB
- Formato:
- Adobe Portable Document Format
- Descripción:
- Acuerdo de publicación
Cargando...
- Nombre:
- 2026PaipaDiego1.pdf
- Tamaño:
- 90.48 KB
- Formato:
- Adobe Portable Document Format
- Descripción:
- Carta de facultad
Bloque de licencias
1 - 1 de 1
Cargando...
- Nombre:
- license.txt
- Tamaño:
- 807 B
- Formato:
- Item-specific license agreed upon to submission
- Descripción:

