Local versus Global Models for Just-In-Time Software Defect Prediction

Algorithm 1

The application process of local models for JIT-SDP.

	Input: training set: test set: the number of clusters: k
	Output: prediction results: R
(1)	begin
(2)
(3)	//Divide the training set into k clusters
(4)	clusters, cluster_centers = k-medoids (D, k)
(5)	//Train a classifier for each cluster
(6)	classifiers = modeling_algorithm (clusters)
(7)	for Instance t in T do
(8)	//Assign a cluster number i for each instance t
(9)	i = calculate_min_distance (t, cluster_centers)
(10)	r = classifiers [i].predict (t) //perform prediction
(11)	R.append (r)
(12)	end
(13)	end