No Detail Found!

Leveraging Machine-Labeled Data and Cross- Lingual Transfer for NER in Urdu and Sindhi

Author(s)
Nazish Basir University of Sindh, Jamshoro Country: Pakistan
Dil Nawaz Hakro University of Sindh, Jamshoro Country: Pakistan
Khalil-Ur-Rehman Khoumbati University of Sindh, Jamshoro Country Pakistan
Zeeshan Bhatti University of Sindh, Jamshoro Country, Pakistan
Abstract
Named Entity Recognition (NER) is a fundamental task in Natural Language Processing (NLP) that involves identifying and classifying entities within text. In low-resource languages like Urdu and Sindhi, due to limited annotated datasets and complex linguistic features such as rich morphology, agglutination, and the absence of capitalization cues the researchers face many challenges. In our study we are introducing a distinct approach that combines machine-labeled data generation with advanced multilingual transformer models to enhance the performance of low resource languages, with cross lingual transfer learning to improve NER performance in Sindhi. To the best of our knowledge, this is the first work exploring cross-lingual NER transfer from Urdu to Sindhi. We have also introduced two new entity types that include colors and foods, which have not been explored previously in Urdu and Sindhi language research. To reduce the need for extensive manual annotation, we used a bagging-based ensemble of Conditional Random Field (CRF) models, to generate high-confidence machinelabeled datasets. These models were trained on subsets of a smaller dataset, which were annotated by language experts. The machinelabeled data notably increased the volume of training data, which is essential for low-resource languages. We pre-trained two models, Multilingual BERT (mBERT) and XLM-RoBERTa on machinelabeled data and fine-tuned them on the human-annotated datasets. Our experiments demonstrated improvements in the performance of the Named Entity Recognition for both languages. Particularly, for Sindhi, the XLM-RoBERTa model's F1 score increased from 0.302 (without pre-training) to 0.681 after pre-training on a combined machine-labeled data of Urdu and Sindhi language, which is approximate increase of 125%. Our results show the effectiveness of incorporating machine-labeled data and cross-lingual knowledge transfer from Urdu to Sindhi language.
Publication Details
Page(s) 1-8
DOI DOI not available
Published Journal: Journal of Information & Communication Technology (JICT), Volume: 19, Issue: 1, Year: 2025

Request the full text from PASTIC

Keywords
mBERT, XLM-RoBERTa, Named Entity Machine-Labeled Data
References
Nadeau D.,Sekine S. . ."A survey of named entity. , : .
Ullah F.,Gelbukh A.,Zamir M. T.,Riverόn E. M. F. . .. , : .
Naz S.,Umar A. I.,Shirazi S. H.,Khan S. A.,I. S. A.,Khan A. A. .2014 .Challenges of Urdu Named Entity Recognition: A Scarce Resourced Language,”. Research Journal of Applied Sciences, Engineering and Technology, 8(10) : 1272-1278.
Jahangir F.,Anwar M. W.,Bajwa U. I.,Wang X. .2012 ."N-gram and Gazetteer List Based Named Entity Recognition for Urdu: A Scarce Resourced Language,". in Proceedings of the 2012 Conference on Language Resources and Evaluation (LREC), : .
Singh A. K. .2008 ."Named entity recognition for South and South East Asian languages: taking stock,". in Proceedings of the IJCNLP-08 Workshop on Named Entity Recognition for South and South East Asian Languages, : .
Khan S. N. .2018 ."Urdu Word Segmentation using Machine Learning Approaches,". International Journal of Advanced Computer Science and Applications, 9(8) : 377-385.
Devlin J.,Chang W.,Lee K.,Toutanova K. .2019 ."BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding,". in Proceedings of NAACL-HLT, : 4171-4186.
Conneau A. .2020 ."Unsupervised Cross-lingual Representation Learning at Scale,". in *Proceedings of ACL, : 8440-8451.
Kim J.,Ko Y.,Seo J. .2020 ."Construction of MachineLabeled Data for Improving Named Entity Recognition by Transfer Learning,". in Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, : 6006-6015.
.2015 .. , 34(4) : 349-358.
Khan W.,Daud A.,Alotaibi F.,Aljohani N.,Arafat S. .2020 ."Deep recurrent neural networks with word embeddings for Urdu named entity recognition,". ETRI Journal, 42(1) : 90-100.
Kazi S.,Rahim M.,Khoja S.,Jan S. .2023 ."A deep learning approach to building a framework for Urdu POS and NER," Intelligence (AI). , 31 : 68.
M. K. Malik .2017 ."Urdu Named Entity Recognition and Classification System Using Artificial Neural Network,". International Journal of Advanced Computer Science and Applications, 8(9) : 139-144.
Ahmed A.,Huang D.,Arafat S. Y.,Apr S. Y. .2024 .Enriching Urdu NER with BERT Embedding, Data Augmentation, and Hybrid Encoder-CNN Architecture,”. ACM Transactions on Asian and Low-Resource Language Information Processing, 23(4) : .
Bahad S.,Mishra P.,Arora K.,Balabantaray R. C.,Sharma D. M.,Krishnamurthy P. .2024 ."Fine-tuning Pre-trained Named Entity Recognition Models For Indian Languages,". arXiv preprint arXiv:2405, : .
Ali W.,Kehar A.,Shaikh H. .2015 ."Towards Sindhi named entity recognition: Challenges and opportunities,". in 1st National Conference on Trends and Innovations in Information Technology, : .
Wang J. .2021 ."Cross-lingual Transfer Learning for LowResource Natural Language Processing Tasks,". Ph.D. dissertation, : .
Hakro D. N.,Hakro M. A.,Lashari and I. A.,Jan and I. A. .2017 ."Sindhi Named Entity Recognition (SNER),". Government: Research Journal of Political Science, 5 : .
Riaz K. .2010 .Rule-based Named Entity Recognition in Urdu,” Association for Computational Linguistics. , : .
Singh U.,Goyal V.,Lehal G. S. .2012 .Named Entity Recognition System for Urdu,”. In Proceedings of COLING, : 2507-2518.
Riaz F.,Anwar M. W.,Muqades H. .2020 ."Maximum entropy based Urdu named entity recognition," in 2020 International Conference on Engineering and Emerging Technologies (ICEET). , : 1-5.
Naz S.,Umar A. I.,M. I. Razzak A. I. . ."A hybrid approach for NER system for scarce resourced languageUrdu: Integrating n-gram with rules and gazetteers,". , : .
A. K. Jumani M. A.,Memon F. H.,Khoso A. A.,Sanjrani S.,Soomro S. .2018 ."Named entity recognition system for Sindhi language," in Emerging Technologies in Computing: First International Conference. , (24) : 237-246.
Ali W.,Lu J.,Xu Z. .2020 ."SiNER: A large dataset for Sindhi named entity recognition," in Proceedings of the Twelfth Language Resources. , : 2953-2961.
Ali W.,Kumar J.,Xu Z.,Kumar R.,Ren Y.,Sep Y. .2021 ."Context-Aware Bidirectional Neural Model for Sindhi Named Entity Recognition,". Applied Sciences, 11(19) : 9038.
Ali W.,Kumar R.,Dai Y.,Kumar J.,Tumrani S. .2021 ."Neural Joint Model for Part-of-Speech Tagging and Entity Extraction,". , : 239-245.
Ali W.,Tumrani S.,Kumar J.,Soomro T. R.,Aug T. R. .2024 ."An Evaluation of Sindhi Word Embedding in Semantic Analogies and Downstream Tasks,". arXiv preprint arXiv:2408, : .
Jean G. .2023 .Cross-Lingual Transfer Learning for LowResource NLP Tasks: Leveraging Multilingual Pretrained Models. , : .
Kadidam V. V. .2024 .Cross Lingual Named Entity Recognition using Deep Learning. , : .
Wang H.,Zhou L.,Duan J.,He L.,Feb L. .2023 .Cross-Lingual Named Entity Recognition Based on Attention and Adversarial Training,”. Applied Sciences (Switzerland), 13(4) : .
Wu Q.,Lin Z.,Karlsson B. F.,Huang B.,Lou J. G.,Jul J. G. .2020 ."Unitrans: Unifying model transfer and data transfer for cross-lingual named entity recognition with unlabeled data,". , : .
Li Z.,Hu C.,Guo X.,Chen J.,Qin W.,Zhang R. .2022 ."An unsupervised multiple-task and multiple-teacher model for cross-lingual named entity recognition," in Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). May, : 170-179.
Zhou R.,Li X.,Bing L.,Cambria E.,Si L.,Miao C.,Nov C. .2022 ."ConNER: Consistency training for cross-lingual named entity recognition,". arXiv preprint arXiv:2211, : .
Kim J.,Ko Y.,Seo J. .2020 .Construction of machinelabeled data for improving named entity recognition by transfer learning'. IEEE Access, 8 : 59684-59693.
Kim J.,Ko Y.,Seo J.,Saeed H.,Qureshi S. .2017 ."Improving Urdu NER by Combining Machine Learning and Rule-Based Approaches,". in Proceedings of the 2017 International Conference on Asian Language Processing (IALP), 8 : 34-37.
0

Citations

0

Downloads

35

Views