Table 5 The optimal parameters of the ML algorithms
Algorithms | Data set 1 | Data set 2 |
|---|---|---|
k-NN | n_neighbors = 13, metric = minkowski | n_neighbors = 19, metric = manhattan |
DT | max_depth = 4, criterion = gini | max_depth = 3, criterion = gini |
LR | C = 4.037, penalty = l1, solver = liblinear | C = 0.498, penalty = l2, solver = newton-cg |
SVM | C = 1.0, gamma = scale, kernel = poly | C = 0.1, gamma = scale, kernel = poly |
MLP | hidden_layer_sizes = 2, max_iter = 1000, solver = lbfgs | hidden_layer_sizes = 16, max_iter = 1000, solver = lbfgs |
GNB | var_smoothing = 0.04 | var_smoothing = 0.196 |