Medina Quero, JavierLopez Medina, Miguel AngelSalguero Hidalgo, AlbertoEspinilla, Macarena2023-02-122023-02-122018-01-012169-3536http://hdl.handle.net/10668/18966Predicting the urgency demand of patients at health centers in smart cities supposes a challenge for adapting emergency service in advance. In this paper, we propose a methodology to predict the number of cases of chronic obstructive pulmonary disease (COPD) from environmental sensors located in the city of Jaen (Spain). The approach presents a general methodology to predict events from environmental sensors within smart cities based on four stages: 1) summarize and expand features by means of temporal aggregations; 2) evaluate the correlation for selecting relevant features; 3) integrate straightforwardly expert knowledge under a fuzzy linguistic approach; and 4) predict the target event with the sequence-based classifier long short-term memory under a sliding window approach. The results show an encouraging performance of the methodology over the COPD patients of the city of Jaen based on a quantitative regression analysis and qualitative categorization of data.enAttribution-NonCommercial-NoDerivatives 4.0 Internationalhttp://creativecommons.org/licenses/by-nc-nd/4.0/Predicting urgency demandlong short-term memorytemporal aggregationfuzzy linguistic approachObstructive pulmonary-diseaseLinguistic term setsHospital admissionsCityPollenresearch articleopen access10.1109/ACCESS.2018.2828652https://doi.org/10.1109/access.2018.2828652433464400001