AlbNER: A Corpus for Named Entity Recognition in Albanian

TL;DR

This paper introduces AlbNER, a new annotated corpus for Albanian named entity recognition, highlighting the importance of language transfer over model size, and providing a resource for future NLP research in Albanian.

Contribution

The creation of AlbNER, a novel annotated corpus for Albanian NER, and an analysis of model performance emphasizing language transfer effects.

Findings

Model size has minimal impact on NER performance.

Language transfer significantly affects NER results.

AlbNER provides a baseline for future Albanian NLP research.

Abstract

Scarcity of resources such as annotated text corpora for under-resourced languages like Albanian is a serious impediment in computational linguistics and natural language processing research. This paper presents AlbNER, a corpus of 900 sentences with labeled named entities, collected from Albanian Wikipedia articles. Preliminary results with BERT and RoBERTa variants fine-tuned and tested with AlbNER data indicate that model size has slight impact on NER performance, whereas language transfer has a significant one. AlbNER corpus and these obtained results should serve as baselines for future experiments.