Table 2 Results comparison.

From: A comparative study of predicting the availability of power line communication nodes using machine learning

Algorithm

Accuracy (%)

F1-Score

Precision

Recall

Training time (sec.)

Memory(MB)

Process CPU Threads

Correctly Classified Instances

0 out of 48

1 out of 52

MLP

84

0.8384

0.8456

0.8373

0.359787

0.523651

25 threads/31.902 sec.

37

47

KNN

67

0.6697

0.6737

0.6723

0.003982

0.539646

21 threads/7.998 sec.

35

32

SVM Linear Kernal

85

0.8465

0.8698

0.8454

0.024401

0.545628

21 threads/2.885 sec.

36

49

SVM Non-Linear Kernal

86

0.8572

0.8769

0.8558

0.043466

0.549362

23 threads/1.775 sec.

36

50

Random Forest

85

0.8465

0.8698

0.8454

0.086674

0.551956

21 threads/2.292 sec.

35

50

ADA Boost

87

0.8661

0.9

0.8646

0.020187

0.553989

25 threads/74.03 sec.

35

52

Back to article page