Software Defect Prediction for Healthcare Big Data: An Empirical Evaluation of Machine Learning Techniques

Table 13

The results of different techniques in terms of accuracy along with the rank values (it ranks the technique for each dataset separately, the best performing algorithm getting the rank of 1 and the second-best rank 2. Last two columns present the sum and average of ranks for each technique.).

	SVM	J48	RF	MLP	RBF	HMM	CDT	A1DE	NB	KNN
AR1	91.73 (2.5)	90.08 (4.25)	90.08 (4.25)	90.08 (4.25)	91.73 (2.5)	92.56 (1.5)	92.56 (1.5)	90.90 (3)	85.12 (5)	90.08 (4.25)
AR3	88.88 (5)	87.30 (6.33)	92.06 (3.5)	93.65 (2)	87.30 (6.33)	97.30 (1)	87.30 (6.33)	92.06 (3.5)	90.47 (4)	85.71 (7)
CM1	89.55 (2.5)	87.95 (5)	89.15 (4)	87.55 (6)	89.55 (2.5)	90.16 (1)	89.35 (3)	86.34 (7)	85.34 (8)	84.39 (9)
JM1	81.74 (4)	79.93 (8)	82.66 (1)	81.96 (3)	82.02 (2)	18.33 (10)	81.65 (5)	81.47 (6)	81.41 (7)	77.08 (9)
KC2	82.75 (6)	81.41 (7)	83.33 (4.5)	84.67 (1)	83.71 (2)	79.50 (9)	82.95 (5)	83.33 (4.5)	83.52 (3)	80.45 (8)
KC3	81.95 (1.5)	79.38 (4)	81.44 (2)	77.31 (7)	79.89 (3.5)	18.55 (9)	81.95 (1.5)	79.89 (3.5)	78.86 (6)	72.16 (8)
MC1	99.28 (5.33)	99.36 (4)	99.48 (1.5)	99.41 (2)	99.28 (5.33)	99.28 (5.33)	99.37 (3)	98.21 (6)	94.15 (7)	99.49 (1.5)
Sum (accuracy)	615.93	605.43	618.23	614.66	613.52	495.70	615.16	612.24	598.90	589.39
Average (accuracy)	87.99	86.49	88.32	87.81	87.65	70.81	87.88	87.46	85.56	84.20
Sum (rank)	26.83	38.58	20.75	25.25	24.16	36.83	25.33	33.50	40.00	46.75
Average (rank)	3.83	5.51	2.96	3.61	3.45	5.26	3.62	4.79	5.71	6.68