Study of Friction and Wear Behavior of Graphene-Reinforced AA7075 Nanocomposites by Machine Learning

Table 2

Wear rate, optimum models.

Type name	Chosen factors
Artificial neural network	Activation function: tan h, α = 0.05, hidden layers = (10,10,10)
k-Nearest neighbor	Number of considered neighbors = 4, weights = “uniform”
Random forest	Maximum features = 6, n_estimators = 30
Support vector machine	Kernel = RBF, γ = 0.3, C = 100
Gradient boosting machine	Maximum depth = 8, learning rate = 0.02, n_estimator = 150

RBF, radial basis function.