Table 1 Hyperparameter optimization for prediction models

\({{\varvec{F}}}_{{\varvec{y}}}\)

–

–

–

max_depth = 9

colsample_bytree = 0.8

min_child_weight = 1

subsample = 0.6

eta = 0.5

\({{\varvec{F}}}_{{\varvec{m}}}\)

n_estimators = 20

max_depth = 10

kernel = "linear"

n_estimators = 750

min_samples_split = 8

max_depth = 1

learning_rate = 0.05

max_depth = 8

colsample_bytree = 0.8

min_child_weight = 1

subsample = 0.8

eta = 0.1

\({{\varvec{K}}}_{{\varvec{e}}}\)

n_estimators = 20

max_depth = 10

kernel = "linear"

–

max_depth = 8

colsample_bytree = 0.8

min_child_weight = 1

subsample = 0.8

eta = 0.4

\({\varvec{D}}\)

–

–

n_estimators = 100

min_samples_split = 2

max_depth = 3

learning_rate = 0.01

max_depth = 5

colsample_bytree = 0.5

min_child_weight = 3

num_boost_round = 10

subsample = 0.5

eta = 0.1