Interval Type-2 Fuzzy Rule-Based System Applications

Six applications of interval type-2 theory, along with their own contributions and explanations, are presented in Chap. 3. The methodology applied to the different models is diverse in terms of mathematical tools. In order of appearance in Sect. 3.1 a fuzzy parameter is introduced in an ordinary differential equation for a pharmacological model; Sect. 3.2 the training of a neural network, ANFIS, for estimating the Area Under the Curve of the Receiver Operating Characteristic for individuals with organ-confined prostate cancer; Sect. 3.3 a p-fuzzy system applied to the evolution of HIV-seropositive population that transfers from asymptomatic to symptomatic, without antiretroviral therapy, using the Mamdani inference method; Sect. 3.4 one Cellular Automaton (CA) to simulate an epidemiological model of the HIV infection, with two inputs provided by the outputs of two interval type-2 FRBS; in the same section, another CA modeling the HIV-seropositive population that changes from asymptomatic to symptomatic and back, with antiretroviral therapy; Sect. 3.5 three other applications of p-fuzzy systems in population models using different inference methods for the type-1 FRBS, which inspire the interval type-2 counterparts; Sect. 3.6 finally a model for the COVID-19 epidemic confirmed numbers of susceptible, infected and, removed (recovered or dead) individuals that uses a fuzzy system identification presented in Chap. 2.