ParsTwiNER: A Corpus for Named Entity Recognition at Informal Persian

MohammadMahdi Aghajani, AliAkbar Badri, Hamid Beigy


Abstract
As a result of unstructured sentences and some misspellings and errors, finding named entities in a noisy environment such as social media takes much more effort. ParsTwiNER contains about 250k tokens, based on standard instructions like MUC-6 or CoNLL 2003, gathered from Persian Twitter. Using Cohen’s Kappa coefficient, the consistency of annotators is 0.95, a high score. In this study, we demonstrate that some state-of-the-art models degrade on these corpora, and trained a new model using parallel transfer learning based on the BERT architecture. Experimental results show that the model works well in informal Persian as well as in formal Persian.
Anthology ID:
2021.wnut-1.16
Volume:
Proceedings of the Seventh Workshop on Noisy User-generated Text (W-NUT 2021)
Month:
November
Year:
2021
Address:
Online
Editors:
Wei Xu, Alan Ritter, Tim Baldwin, Afshin Rahimi
Venue:
WNUT
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
131–136
Language:
URL:
https://aclanthology.org/2021.wnut-1.16
DOI:
10.18653/v1/2021.wnut-1.16
Bibkey:
Cite (ACL):
MohammadMahdi Aghajani, AliAkbar Badri, and Hamid Beigy. 2021. ParsTwiNER: A Corpus for Named Entity Recognition at Informal Persian. In Proceedings of the Seventh Workshop on Noisy User-generated Text (W-NUT 2021), pages 131–136, Online. Association for Computational Linguistics.
Cite (Informal):
ParsTwiNER: A Corpus for Named Entity Recognition at Informal Persian (Aghajani et al., WNUT 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.wnut-1.16.pdf
Code
 overfit-ir/parstwiner
Data
ParsTwinerCoNLL 2003PEYMA