Table 2 The hyper-parameter settings of the machine learning models

ANN

hidden;

linear.output

hidden:c(1,1,1) ~ c(10,10,10);

linear.output:FALSE;

learningrate: 0.001 ~ 0.050

hidden = c(10,7,4);

linear.output = FALSE

learningrate = 0.013

DT

method;

metric;

trControl

tuneGrid

method: “C5.0”;

metric: “ROC”;

trControl: trainControl(method = “cv”,selectionFunction = “oneSE”)

tuneGrid* = expand.grid(.model = “tree”,.trials = c(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40))

method = “C5.0”;

metric = “ROC”;

trControl = trainControl(method = “cv”,selectionFunction = “oneSE”)

tuneGrid* = expand.grid(.model = “tree”,.trials = c(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40))

KNN

kernel

Kernel: “triweight”;

k: 1 ~ 30;

d: 1 ~ 10

kernel = “triweight”;

k = 15;

d = 2

LR

maxit

maxit: 1 ~ 100

maxit = 50

NB

-

-

-

RF

method;

selectionFunction;

metric;

trControl

method: “rf”;

selectionFunction: “oneSE”;

metric: “Kappa”;

trControl:trainControl(method = “cv”,selectionFunction = “oneSE”)

method = “rf”;

selectionFunction = “oneSE”;

metric = “Kappa”;

trControl = trainControl(method = “cv”,selectionFunction = “oneSE”)

SVM

method;

tuneLength;

trControl

method: “svmRadial”;

tuneLength: 1 ~ 50;

trControl:trainControl(method = “cv”, selectionFunction = “oneSE”)

method = “svmRadial”;

tuneLength = 12;

trControl = trainControl(method = “cv”,selectionFunction = “oneSE”,classProbs = TRUE)