Fuzzy gradual rules in epidemiology

Abstract

This paper presents an application of the fuzzy gradual rules in an epidemic study of canine rabies in São Paulo city, Brazil. A linguistic epidemiological model was elaborated through fuzzy rules built by the Extension Principle. We used both the inference method of Mamdani and of Dubois et al. The results were compared with real data from São Paulo and with another MISO Model, which is entirely based on expert knowledge presented in a previous work. Questions about application of fuzzy techniques in epidemiology, different inference methods and the Dubois et al. methodology are discussed.