Country-level Arabic Dialect Identification using RNNs with and without Linguistic Features

Elsayed Issa, Mohammed AlShakhori1, Reda Al-Bahrani, Gus Hahn-Powell


Abstract
This work investigates the value of augmenting recurrent neural networks with feature engineering for the Second Nuanced Arabic Dialect Identification (NADI) Subtask 1.2: Country-level DA identification. We compare the performance of a simple word-level LSTM using pretrained embeddings with one enhanced using feature embeddings for engineered linguistic features. Our results show that the addition of explicit features to the LSTM is detrimental to performance. We attribute this performance loss to the bivalency of some linguistic items in some text, ubiquity of topics, and participant mobility.
Anthology ID:
2021.wanlp-1.32
Volume:
Proceedings of the Sixth Arabic Natural Language Processing Workshop
Month:
April
Year:
2021
Address:
Kyiv, Ukraine (Virtual)
Venue:
WANLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
276–281
Language:
URL:
https://aclanthology.org/2021.wanlp-1.32
DOI:
Bibkey:
Cite (ACL):
Elsayed Issa, Mohammed AlShakhori1, Reda Al-Bahrani, and Gus Hahn-Powell. 2021. Country-level Arabic Dialect Identification using RNNs with and without Linguistic Features. In Proceedings of the Sixth Arabic Natural Language Processing Workshop, pages 276–281, Kyiv, Ukraine (Virtual). Association for Computational Linguistics.
Cite (Informal):
Country-level Arabic Dialect Identification using RNNs with and without Linguistic Features (Issa et al., WANLP 2021)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2021.wanlp-1.32.pdf