Using XGBoost to Discover Infected Hosts Based on HTTP Traffic

Table 5

Experimental parameters settings.

Parameter	Description	Value
booster	Tree model	gbtree
gamma	For pruning	1
max_depth	Depth of decision tree	12
scale_pos_weight	Balance positive and negative sample weights	1
subsample, colsample_bytree	Proportion of random collected samples each time	1
min_child_weight	Number of leaf nodes in the decision tree	1000
eta	Learning rate	0.1