Software Defect Prediction through Neural Network and Feature Selections

Table 7

Comparison of RBF results with other previous methods before feature selection for JM1 data set.

Source	Algorithm	F-measure	Accuracy
[31]	RF	0.76	77.00%
	DS	0.71	71.00%
	SVM	0.71	69.00%
[47]	Naive Bayes	0.89	81.43%
	MLP	0.90	89.97%
	SVM	0.90	81.73%
	RBF	0.90	81.61%
[42]	J48	N/A	79.81%
This research	RBF	0.87	84.87