Table 4 Performance comparison of DL models with the proposed model.

From: Hybrid pre trained model based feature extraction for enhanced indoor scene classification in federated learning environments

Feature Extractor

Classifier

Accuracy (%)

Precision (%)

Recall (%)

F1-Score (0–1)

ZOL

(0–1)

Methews Coefficient

Kappa Statistics

VGG16

Decision Tree

52

53

52

0.52

0.481433

0.504781

0.504561

Logistic Regression

89

92

91

0.91

0.107471

0.889452

0.8894

Random Forest

82

91

82

0.86

0.175623

0.819558

0.818744

Extra Trees

82

92

81

0.86

0.177807

0.817684

0.816322

Hist Gradient Boosting

86

91

86

0.88

0.139799

0.856322

0.855894

Multi Layered Perceptron

63

63

60

0.61

0.371341

0.617851

0.61765

VGG19

Decision Tree

51

51

50

0.5

0.492355

0.493621

0.493427

Logistic Regression

88

90

89

0.89

0.118829

0.877794

0.877713

Random Forest

79

88

78

0.82

0.207951

0.786128

0.785316

Extra Trees

80

90

77

0.82

0.203145

0.791253

0.790073

Hist Gradient Boosting

23

27

26

0.23

0.773263

0.210248

0.207227

Multi Layered Perceptron

64

64

59

0.6

0.36173

0.627556

0.627236

ResNet 152

Decision Tree

28

26

26

0.26

0.721979

0.25679

0.256733

Logistic Regression

67

71

69

0.7

0.327058

0.663184

0.66305

Random Forest

56

74

51

0.57

0.440455

0.544877

0.542807

Extra Trees

55

75

48

0.55

0.453152

0.531971

0.529031

Hist Gradient Boosting

61

75

58

0.63

0.388792

0.598489

0.596998

Multi Layered Perceptron

42

41

36

0.37

0.578371

0.402491

0.401788

InceptionResNet

Decision Tree

25

22

22

0.22

0.74574

0.233011

0.232868

Logistic Regression

65

69

67

0.68

0.349061

0.6406

0.640503

Random Forest

51

66

44

0.49

0.487112

0.496484

0.49406

Extra Trees

50

70

43

0.49

0.503713

0.479194

0.475773

Hist Gradient Boosting

50

40

20

0.01

0.95107

0.023288

0.008262

Multi Layered Perceptron

40

36

32

0.32

0.60332

0.376451

0.375507

Mobile Net

Decision Tree

21

19

18

0.18

0.790738

0.187306

0.187208

Logistic Regression

56

58

57

0.57

0.442551

0.544893

0.544743

Random Forest

44

59

37

0.41

0.563565

0.41797

0.415247

Extra Trees

42

61

36

0.4

0.576234

0.404824

0.401376

Hist Gradient Boosting

48

63

44

0.49

0.515946

0.467625

0.465104

Multi Layered Perceptron

35

33

28

0.29

0.649192

0.329995

0.329104

Dense Net

Decision Tree

34

32

31

0.31

0.656743

0.324706

0.324571

Logistic Regression

77

80

78

0.79

0.231173

0.762368

0.762249

Random Forest

64

79

59

0.64

0.359457

0.630222

0.628265

Extra Trees

63

80

58

0.64

0.366025

0.62359

0.621119

Hist Gradient Boosting

90

16

60

0.06

0.911559

0.068907

0.050004

Multi Layered Perceptron

49

45

43

0.43

0.507881

0.476921

0.476427

Multi Data

Decision Tree

99

99

98

0.99

0.009632

0.99009

0.990083

Logistic Regression

100

100

100

1

0.001313

0.998648

0.998648

Random Forest

100

100

100

1

0.000876

0.999099

0.999099

Extra Trees

100

100

100

1

0.000438

0.999549

0.999549

Hist Gradient Boosting

100

100

100

1

0.003065

0.996846

0.996845

Multi-Layered Perceptron

98

97

97

0.97

0.0162

0.983429

0.983323

Back to article page