Named Entity Recognition for the Kurdish Sorani Language: Dataset Creation and Comparative Analysis

TL;DR

This paper introduces the first named entity recognition dataset for Kurdish Sorani, compares classical and neural models, and finds that traditional methods outperform neural approaches in low-resource settings.

Contribution

It creates the first NER dataset for Kurdish Sorani and provides a comparative analysis showing classical models can outperform neural ones in low-resource NLP tasks.

Findings

CRF achieves an F1-score of 0.825

BiLSTM models score 0.706

Classical methods outperform neural models in this context

Abstract

This work contributes towards balancing the inclusivity and global applicability of natural language processing techniques by proposing the first 'name entity recognition' dataset for Kurdish Sorani, a low-resource and under-represented language, that consists of 64,563 annotated tokens. It also provides a tool for facilitating this task in this and many other languages and performs a thorough comparative analysis, including classic machine learning models and neural systems. The results obtained challenge established assumptions about the advantage of neural approaches within the context of NLP. Conventional methods, in particular CRF, obtain F1-scores of 0.825, outperforming the results of BiLSTM-based models (0.706) significantly. These findings indicate that simpler and more computationally efficient classical frameworks can outperform neural architectures in low-resource settings.