Cache Pollution Detection Method Based on GBDT in Information-Centric Network

Table 5

LightGBM-related parameter settings.

Parameter	Value	Description
Objective	Binary	Two classification problem
Num_leaves	50	Up to 50 leaf nodes per tree
max_depth	8	The maximum depth of the tree is 8
learning_rate	0.01	Learning rate is 0.01
reg_alpha	0.005	L1 regular weight is 0.005
reg_lambda	0.0005	L2 regular weight 0.0005
Subsample	0.9	Selective feature probability
n_estimators	4000	Maximum number of iterations
early_stopping_rounds	200	Early stopping rounds