Prediction of Defective Software Modules Using Class Imbalance Learning

Table 5

Comparison on the basis of precision values on eight datasets.

Dataset	SVM	CBNN	NB	RF	LR	-NN	BBN	C4.5	LSTSVM	WLSTSVM
CM1	0.2562	0.3854	0.4038	0.3589	0.3464	0.4200	0.5203	0.3420	0.4225	0.8685
KC1	0.4468	0.5203	0.4797	0.4282	0.5402	0.6436	0.4953	0.2975	0.9538	0.9475
PC1	0.5538	0.4953	0.4861	0.4566	0.3536	0.4593	0.6516	0.4321	0.9532	0.9792
PC3	0.6027	0.6228	0.4176	0.3805	0.4205	0.3051	0.5164	0.5163	0.4287	0.6873
PC4	0.5166	0.6512	0.5648	0.6250	0.5466	0.5501	0.5278	0.4833	0.5310	0.8667
MC2	0.5632	0.7775	0.6185	0.6393	0.4839	0.4499	0.5047	0.5384	0.7504	0.8824
KC2	0.5883	0.8674	0.5766	0.6488	0.7063	0.6207	0.7212	0.6224	0.9541	0.9326
KC3	0.4054	0.6484	0.5237	0.5466	0.5321	0.5293	0.6653	0.2991	0.5357	0.6157