You should develop your fuzzy systems jointly within your Lab Group in either R$,$ Python or Java. You can

build on your own fuzzy system implementation as developed during lab sheets $1$ and $2.$ You can also make

use of the fuzzy system libraries $.$ Note that not all libraries may

provide all the features you are looking for. Where this is the case, you should extend the functionality

provided using your own code.

You should construct two fuzzy systems designed specifically to work for the following cases:

Case $1$: All inputs are numeric, i$.$e$.$ Temperature, Headache and Age are all provided as numbers.

Case $2$: All inputs are interval$-$valued, i$.$e$.$ Temperature, Headache and Age are provided as intervals.

When constructing the FIS models, you should carefully consider and justify:

$$ the number of terms in each input variable and in the output variable, giving each term an

appropriate linguistic label

$$ the types of membership functions $($triangular$,$ Gaussian, hybrid, data$-$driven, etc.$)$

$$ the parameters of each membership function $($respectively the underlying data$)$

$$ the fuzzification approach

$$ the rules that connect the input terms to the output terms

$$ the operators used in the FIS $($consider both min$/$max and product$/$probabilistic$-$or families$)$

$$ various defuzzification methods including numeric and linguistic

Note:

$$ your systems will not be evaluated for medical accuracy in any way. However, their function should

be sensible and you should justify how$/$why it makes sense

$$ you should develop the most appropriate type of fuzzy system for Case $1$ and Case $2,$ using your

report to discuss how inputs are processed differently and what the impact is on the outputs

generated

$$ you may develop more than two systems in order to compare and contrast for example different

parameter selections or approaches, however the same page limit applies.