Data Mining based

Techniques

Correctly Classified Instances (%)

Incorrectly Classified Instances (%)

TP (True Positive) Rate

FP (False Positive) Rate

Precision

Recall

F1 Score

Experiment I (Variable length)

100.00%

0.00%

1

0

1

1

1

Experiment II (Variable length)

0.00%

100.00%

0

1

0

0

0

Experiment III (Equal length)

100.00%

0.00%

1

0

1

1

1

Chen et al. [16] ―J48

before alignment

Training

85.00%

15.00%

-

-

-

-

-

5-fold cross validation

60.00%

40.00%

-

-

-

-

-

10-fold cross validation

63.33%

36.67%

-

-

-

-

-

15-fold cross validation

68.33%

31.67%

-

-

-

-

-

20-fold cross validation

60.00%

40.00%

-

-

-

-

-

Chen et al. [16] ―J48

after double alignment

Training

96.67%

3.33%

-

-

-

-

-

5-fold cross validation

78.33%

21.67%

-

-

-

-

-

10-fold cross validation

66.67%

33.33%

-

-

-

-

-

15-fold cross validation

70.00%

30.00%

-

-

-

-

-

20-fold cross validation

63.33%

36.67%

-

-

-

-

-

Kumar et al. [44]

Existing (known)dataset (Average)

95.9752%

4.0248%

0.96

0.094

0.962

0.96

0.959

New (unknown)dataset (Average)

86.6873%

13.3127%

0.867

0.275

0.872

0.867

0.858

Prabha et al. [45]

-

-

-

-

-

-

-

-

Statistical method by Srakaew et al. [18]

Reference Set

98.9167%

1.0833%

-

-

-

-

-

Application Set

95.0477%

4.9523%

-

-

-

-

-

10-fold cross validation

95.333%

4.667%

-

-

-

-

-

Abstract assembly method by Srakaew et al. [18]

Reference Set

99.75%

0.25%

-

-

-

-

-

Application Set

98.39%

1.661%

-

-

-

-

-

10-fold cross validation

99.5%

0.5%

-

-

-

-

-