Predicting Subcontractor Performance Using Web-Based Evolutionary Fuzzy Neural Networks

Table 2

EFNN parameters.

Technique	Parameter	Value
GAs	Population size	50
	Crossover rate	0.5
	Mutation rate	0.01
	Terminal condition	5000 iterations
FL	Number of fuzzy sets	5
	Defuzzification function	Average output method
	MF shape	Trapezoidal shape
NNs	Connection weight	0.0–1.0
	Bias	−1–0
	Activation function	Heaviside
	Hidden layers	1–6
	Hidden neurons	1–6