eprintid: 2118 rev_number: 13 eprint_status: archive userid: 2 dir: disk0/00/00/21/18 datestamp: 2022-05-31 18:14:34 lastmod: 2023-07-11 23:30:12 status_changed: 2022-05-31 18:14:34 type: article metadata_visibility: show creators_name: Mahajan, Asmita creators_name: Sharma, Nonita creators_name: Aparicio Obregón, Silvia creators_name: Alyami, Hashem creators_name: Alharbi, Abdullah creators_name: Anand, Divya creators_name: Sharma, Manish creators_name: Goyal, Nitin creators_id: creators_id: creators_id: silvia.aparicio@uneatlantico.es creators_id: creators_id: creators_id: divya.anand@uneatlantico.es creators_id: creators_id: title: A Novel Stacking-Based Deterministic Ensemble Model for Infectious Disease Prediction ispublished: pub subjects: uneat_eng divisions: uneatlantico_produccion_cientifica divisions: unic_produccion_cientifica full_text_status: public keywords: autoregressive integrated moving average; epidemiology; exponential smoothing; ensemble; gradient boosting; infectious disease; neural network autoregression; pandemic; stacking abstract: Infectious Disease Prediction aims to anticipate the aspects of both seasonal epidemics and future pandemics. However, a single model will most likely not capture all the dataset’s patterns and qualities. Ensemble learning combines multiple models to obtain a single prediction that uses the qualities of each model. This study aims to develop a stacked ensemble model to accurately predict the future occurrences of infectious diseases viewed at some point in time as epidemics, namely, dengue, influenza, and tuberculosis. The main objective is to enhance the prediction performance of the proposed model by reducing prediction errors. Autoregressive integrated moving average, exponential smoothing, and neural network autoregression are applied to the disease dataset individually. The gradient boosting model combines the regress values of the above three statistical models to obtain an ensemble model. The results conclude that the forecasting precision of the proposed stacked ensemble model is better than that of the standard gradient boosting model. The ensemble model reduces the prediction errors, root-mean-square error, for the dengue, influenza, and tuberculosis dataset by approximately 30%, 24%, and 25%, respectively date: 2022-05 date_type: published publication: Mathematics volume: 10 number: 10 pagerange: 1714 id_number: doi:10.3390/math10101714 refereed: TRUE issn: 2227-7390 official_url: http://doi.org/10.3390/math10101714 access: open language: en citation: Artículo Materias > Ingeniería Universidad Europea del Atlántico > Investigación > Producción Científica Universidad Internacional do Cuanza > Investigación > Producción Científica Abierto Inglés Infectious Disease Prediction aims to anticipate the aspects of both seasonal epidemics and future pandemics. However, a single model will most likely not capture all the dataset’s patterns and qualities. Ensemble learning combines multiple models to obtain a single prediction that uses the qualities of each model. This study aims to develop a stacked ensemble model to accurately predict the future occurrences of infectious diseases viewed at some point in time as epidemics, namely, dengue, influenza, and tuberculosis. The main objective is to enhance the prediction performance of the proposed model by reducing prediction errors. Autoregressive integrated moving average, exponential smoothing, and neural network autoregression are applied to the disease dataset individually. The gradient boosting model combines the regress values of the above three statistical models to obtain an ensemble model. The results conclude that the forecasting precision of the proposed stacked ensemble model is better than that of the standard gradient boosting model. The ensemble model reduces the prediction errors, root-mean-square error, for the dengue, influenza, and tuberculosis dataset by approximately 30%, 24%, and 25%, respectively metadata Mahajan, Asmita; Sharma, Nonita; Aparicio Obregón, Silvia; Alyami, Hashem; Alharbi, Abdullah; Anand, Divya; Sharma, Manish y Goyal, Nitin mail SIN ESPECIFICAR, SIN ESPECIFICAR, silvia.aparicio@uneatlantico.es, SIN ESPECIFICAR, SIN ESPECIFICAR, divya.anand@uneatlantico.es, SIN ESPECIFICAR, SIN ESPECIFICAR (2022) A Novel Stacking-Based Deterministic Ensemble Model for Infectious Disease Prediction. Mathematics, 10 (10). p. 1714. ISSN 2227-7390 document_url: http://repositorio.unic.co.ao/id/eprint/2118/1/mathematics-10-01714-v2.pdf