A Software Defect Prediction Approach Based on Hybrid Feature Dimensionality Reduction

Table 2

Experimental data.

Dataset	Project	#Feature	#Instances	#Defect	Defect (%)
AEEEM	EQ	61	324	129	39.81
	JDT	61	997	206	20.66
	LC	61	691	64	9.26
	ML	61	1,862	245	13.16
	PDE	61	1,497	209	13.96
SOFTLAB	AR1	29	121	9	7.44
	AR3	29	63	8	12.70
	AR4	29	107	20	18.69
	AR5	29	36	8	22.22
	AR6	29	101	15	14.85
MORPH	Ant	20	125	20	16
	Camel	20	339	13	3.83
	Poi	20	237	141	59.49
	Tomcat	20	858	77	8.97
	Velocity	20	196	147	75
	Xalan	20	723	110	15.21
	Xerces	20	440	71	16.14
	Arc	20	234	27	11.54
	Redktor	20	176	27	15.34
ReLink	Apache	26	194	98	50.52
	Safe	26	56	22	39.29
	ZXing	26	399	118	29.57