Identification System based on Fuzzy Logic for epidemiological control of dengue in the metropolitan region of Sao Luís - MA

Resumo

Dengue, as a highly prevalent disease in Brazil, impairs the population’s well-being, affecting public
health costs, so it is necessary to take measures to help prevent the onset of the disease. Thus, predicting the number
of cases of the disease in a given region helps in planning and decision-making by Public Agencies. For this reason,
it is extremely important that such forecasts are accurate, although this process has errors, since the factors that
serve to obtain the diagnosis of cases have a behavior that depends on numerous parameters, such as precipitation
rate and ambient temperature . Therefore, fuzzy logic presents itself as a good alternative for modeling a dengue

case prediction system, as well as the development of low-cost technologies for this purpose. For that, a Takagi-
Sugeno (TS) type MISO system (multiple inputs and one output) is proposed, capable of providing a forecast of the

number of dengue cases in the city of Sao Lu ̃ ́ıs - MA, based on the collection of epidemiological data from dengue
in the municipality, which were related to the input variables of the system, obtaining linear submodels through a
fuzzy clustering algorithm. The model obtained, with two input variables (rainfall rate and average temperature),
obtained good computational results.