Cross-Project Defect Prediction Based on Two-Phase Feature Importance Amplification

Algorithm 2

Random Forest.

(i)	Input: , source project dataset; , sample from source project dataset
	Output:  = classification of
(1)	for do
(2)	use Bootstrap on to get the training dataset
(3)	use to generate a tree without pruning
(4)	randomly select features from 's features
(5)	calculate the metric based on (2) at each node to select the optimal features based on the selected features
(6)	Splitting until the tree grows to its maximum
(7)	end for
(8)	return