Skip to main content

Table 2 Algorithms grouping

From: Systematic literature review of the techniques for household electrical appliance anomaly detections and knowledge extractions

Neural networks

No. of papers

Implementations (%)

Traditional classification

No. of papers

Implementations (%)

Convolutional neural networks (CNN)

23

25

Support vector machine (SVM)

37

40.2

Deep autoencoder (DAE)

16

17.4

k-nearest neighbors (KNN)

42

45.7

Deep belief network (DBN)

12

13.1

Logistic regression

13

14.1

Generative adversarial networks (GAN)

3

3.3

   

Recurrent neural network (RNN)

16

17.4

   

Multi-layer perceptron (MLP)

14

15

   

Radial basis functional neural network (RBFNN)

8

8.7

   

Total

92

99.9

  

100

Regression

No. of papers

Implementations (%)

Probabilistic model

No. of papers

Implementations (%)

Support vector regression (SVR)

27

29.3

Naïve bayes

38

41.3

Autoregressive integrated moving average (ARIMA)

56

60.9

Bayesian networks

54

58.7

Autoregression

9

9.8

   

Hybrids

No. of papers

Implementations (%)

Clustering

No. of papers

Implementations (%)

Semi-Support vector machine (semi-SVM)

13

14.13

K-means

69

75

Deep autoencoder-k neural networks graph (DAE-KNNG)

8

8.70

C-means

11

11.96

   

Mutual k-nearest neighbors (MNN)

3

3.26

   

Entropy-based

9

9.78

One-class learning

No. of papers

Level of implementations (%)

Dimensionality reduction

No. of papers

Level of implementations (%)

One-class random forest (OCRF)

4

4.34

Principal component analysis (PCA)

17

18.48

One-class support vector machine (OCSVM)

3

3.26

Linear discriminant analysis (LDA)

11

11.96

One-class neural network (OCNN)

2

5.43

Quadratic discriminant analysis (QDA)

9

9.78

   

Multiple discriminant analysis (MDA)

7

7.61

One-class convolutional neural network (OCCNN)

9

9.78