Neural Fuzzy System
Neural Fuzzy System
Explanation
Temperature and speed of fan form initial inputs whose memberships found by triangular regions .
These memberships form input to a second layer that has rules .Outcome of the rules specify the
fuzzy outputs that correspond to the next layer .Outputs of the penultimate(second-to-last) layer are
defuzzified to give crisp output.
The weights in the neural network change during the training phase.
Training is conducted by presenting both the input and the desired target output.
System’s output is computed and the error between this and desired output is propagated
backwards(using back propagation algorithm)
Activation functions of neurons modified.
After several stages, weights are altered to reveal best values.
During this process some arc’s weight = 0 in which the system filters this unnecessary or
redundant rules that included by the expert unknowingly.
Models of NFS
Applications of NFS