Skip to main content

Table 4 The result of behavior identification

From: Graph representation learning-based residential electricity behavior identification and energy management

Model

Method

B1

B3

H1

H2

H3

H4

H5

H6

H7

H8

H9

H10

MLKNN

HL

0.024

0.093

0.139

0.059

0.118

0.109

0.174

0.158

0.085

0.131

0.049

0.151

 

Acc

0.976

0.906

0.861

0.941

0.882

0.891

0.826

0.842

0.915

0.869

0.951

0.849

 

F1-Score

0.825

0.805

0.467

0.568

0.588

0.604

0.415

0.528

0.589

0.527

0.72

0.693

RAKEL

HL

0.02

0.069

0.142

0.06

0.119

0.11

0.171

0.143

0.081

0.129

0.062

0.144

 

Acc

0.98

0.931

0.858

0.94

0.881

0.89

0.829

0.857

0.919

0.871

0.938

0.855

 

F1-Score

0.84

0.857

0.181

0.54

0.576

0.591

0.4

0.506

0.586

0.502

0.601

0.653

SGN

HL

0.011

0.032

0.078

0.043

0.081

0.075

0.13

0.135

0.05

0.107

0.025

0.131

 

Acc

0.989

0.97

0.922

0.957

0.919

0.925

0.87

0.865

0.95

0.893

0.975

0.869

 

F1-Score

0.918

0.939

0.698

0.69

0.718

0.72

0.511

0.558

0.765

0.64

0.861

0.728

LDWA

HL

0.011

0.030

0.072

0.041

0.078

0.077

0.117

0.133

0.052

0.101

0.021

0.131

 

Acc

0.989

0.97

0.930

0.959

0.922

0.923

0.881

0.865

0.948

0.901

0.975

0.869

 

F1-Score

0.918

0.939

0.733

0.706

0.733

0.723

0.537

0.563

0.769

0.653

0.876

0.728

ML-SGN (proposed)

HL

0.011

0.033

0.065

0.041

0.072

0.073

0.102

0.124

0.048

0.097

0.02

0.131

 

Acc

0.989

0.967

0.935

0.959

0.928

0.927

0.898

0.876

0.952

0.903

0.98

0.869

 

F1-Score

0.92

0.937

0.781

0.72

0.752

0.719

0.574

0.584

0.777

0.684

0.886

0.726