Software Defect Prediction through Neural Network and Feature Selections

Table 18

Comparison of RBF results with other previous methods before feature selection for PC4 data set.

Source	Algorithm	F-measure	Accuracy (%)
[47]	Naive Bayes	0.92	85.51
	MLP	0.94	89.11
	SVM	0.94	88.45
	RBF	0.93	27.27
[31]	RF	0.90	90.00
	DS	0.86	85.00
	SVM	0.84	81.00
	LR	0.84	82.00
[49]	RBF	0.25	87.40
	KNN	0.28	85.82
	DT	0.58	86.87
This research	RBF	0.95	94.44