Correction to: Development and evaluation of an Urdu treebank (CLE-UTB) and a statistical parser
- PDF / 166,704 Bytes
- 2 Pages / 439.37 x 666.142 pts Page_size
- 106 Downloads / 204 Views
Correction to: Development and evaluation of an Urdu treebank (CLE-UTB) and a statistical parser Toqeer Ehsan1
•
Sarmad Hussain2
Ó Springer Nature B.V. 2020
Correction to: Lang Resources & Evaluation https://doi.org/10.1007/s10579-020-09492-7 In the original publication of the article the column headers of the Tables 17 and 18 were incorrectly published. The corrected version of Tables 17 and 18 are provided with this Correction. The original article has been corrected.
The original article can be found online at https://doi.org/10.1007/s10579-020-09492-7. & Toqeer Ehsan [email protected] Sarmad Hussain [email protected] 1
Department of Computer Science and Engineering, University of Engineering and Technology, Lahore, Pakistan
2
Center for Language Engineering (CLE), KICS/CS&E, University of Engineering and Technology, Lahore, Pakistan
123
Lang Resources & Evaluation Table 17 No. of sentences in train and test sets. Test set has been divided into three classes i.e. small, medium and long sentences Dataset
Train
Test
Total
Small (B 10) #Sentences
6881
#Tokens
130,105
Medium (11–25)
Long (> 25)
All test
229
526
218
973
7854
1919
9136
7415
18,470
148,575
Table 18 Parsing results after training two parsers on the CLE-UTB CLE-UTB Version
Gold POS
Predicted POS
Labeled recall (%)
Labeled precision (%)
Labeled f-score (%)
Labeled recall (%)
Labeled precision (%)
Labeled f-score (%)
POS accuracy (%)
Stanford Parser
79.0
77.9
78.4
75.1
74.3
74.7
94.9
BiLSTM Parser
87.1
89.2
88.1
85.7
86.7
86.2
96.3
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
123
Data Loading...
