Table 7 Performance of the GRU-based classifier.

From: A deep learning framework for Ethiopian sign language recognition using skeleton-based representation

Label

Precision

Recall

F1-Score

Support

ሀዘን

0.88

0.93

0.90

30

ህመም

1.00

1.00

1.00

30

ለምን?

0.88

0.96

0.92

23

ልምምድ

1.00

0.85

0.92

27

መራራ

1.00

0.88

0.94

26

መናደድ

0.97

0.91

0.94

32

መጀመር

0.90

1.00

0.95

19

መገረም

0.96

1.00

0.98

24

መጥፎ

0.93

0.93

0.93

29

መጨረስ

0.96

1.00

0.98

23

መጨነቅ

0.94

0.94

0.94

34

ማሸነፍ

1.00

0.84

0.91

25

ምስጋና

0.97

0.97

0.97

34

ቀላል

0.90

1.00

0.95

28

ተነስ!

0.86

0.89

0.88

28

አስቸጋሪ

0.88

1.00

0.94

30

እህት

0.89

0.94

0.92

36

ወንድም

0.95

0.91

0.93

23

ይቅርታ

1.00

0.97

0.98

31

ጎበዝ

0.96

0.86

0.91

28

Accuracy

  

0.94

560

Macro Avg

0.94

0.94

0.94

560

Weighted Avg

0.94

0.94

0.94

560