Table 4 Performance analysis of used algorithms on diabetes classification problem.

From: A hybrid bio inspired neural model based on Ropalidia Marginata behavior for multi disease classification

 

Algorithms

Accuracy

MSE

SD

65

ABCNN

71.88

0.2505

0.0154

CSNN

73.87

0.1505

0.0554

ERN

72.92

0.2708

0.0408

LM

72.92

0.7208

0.0200

ABC-LM

65.09

0.14

0.0330

ABCNN

68.09

0.131

0.0210

69

BMNABC + C4.5

76.17

–

–

BMNABC + KNN

70.44

–

–

BMNABC + NB

76.43

–

–

BMNABC + ODF

77.21

–

–

70

PCA + Naïve Bayes

79.13

–

–

69

Mean imputation + LSTM

85.00

–

–

71

Mean imputation + RB-Bayes

72.90

–

–

72

Mean imputation + NB

76.30

–

–

19

CAPSO-MLP

74.68

0.204

–

PSO-MLP

74.03

0.205

–

GSA-MLP

56.49

0.267

–

ICA-MLP

66.23

0.222

–

66

bSCWDTO-KNN

65.00

3.500

0.2560

bDTO-KNN

63.37

3.663

0.2701

bSC-KNN

64.50

3.550

0.2752

bPSO-KNN

62.68

3.732

0.2593

bWOA-KNN

61.67

3.833

0.2650

bGWO-KNN

65.09

3.491

0.2577

bMVO-KNN

60.03

3.997

0.2560

bSBO-KNN

62.09

3.791

0.2790

bGA-KNN

63.47

3.653

0.2775

bFA-KNN

62.74

3.726

0.2652

bGWO_GA-KNN

65.74

3.426

0.2655

67

bGWDTO-KNN

87.23

0.256

0.1475

68

bGWDTO-KNN

75.64

0.5825

0.4407

Proposed

RMONN

88.38

0.1235

0.0313

RMOBPERN

70.31

0.3073

0.0021

RMOLMBP

97.20

0.049

0.0011

RMOLM

71.88

0.2812

0.0023

Back to article page