In the case of using GBDT, prediction accuracy was considerd to have deteriorated by this extrapolation.
Feature of each prediction method No Method Feature 1 GBDT A machine learning technique for regression and classification problems, which produce a prediction model in the form of an ensemble of weak prediction models, typically decision trees.
(3) Training a prediction model with GBDT optimized by PSO
The goal of GBDT is to learn an optimal model [F.sup.*](x) such that [[summation].sup.n.sub.i=1]L([y.sub.i], F([x.sub.i])) is minimized for a specified loss function L(y, F(x)).
Bai, Jianjun Cheng, and Xiaoyun Chen, "Towards Effective Network Intrusion Detection: A Hybrid Model Integrating Gini Index and GBDT with PSO," Journal of Sensors, vol.
Classifier KNN GNB SVM GBDT Detection time (s) 2888.12 1.42 36939.30 171.95 Detection rate 0.9332 0.8328 0.8992 0.9355 Table 5: Weights and accuracy.
Separately, GBDT, decision trees as the weak learner capable of calculating the feature importance, are applied for comparison with the GAFS schemes.
GBDT was trained with the training data without using mRMR and also evaluated with Cohen's kappa of 10fold CV.
The performance of our proposed CHDNet incorporating LIBLINEAR SVM  was also compared with other classifiers, including LDA, KNN, CART, GBDT
, RF, and LIBSVM using identical data and features.
SVM, RF, and GBDT
keep relatively stable f-measures for all activities.
As can be seen from Figure 4, GBDT
is a kind of linear training, so it cannot be trained in parallel.
Significantly different from the random forest, the tree-based models of GBDT
are trained sequentially, and each base model is added to correct the error produced by its previous tree models.