NERCat: Fine-Tuning for Enhanced Named Entity Recognition in Catalan

TL;DR

The paper tackles the challenge of Named Entity Recognition in Catalan, a low-resource language, by fine-tuning GLiNER on a high-quality, manually annotated dataset derived from Catalan TV transcripts. It combines a careful dataset construction, manual annotation, and targeted fine-tuning of the GLiNER Knowledgator model, supplemented by real-world transcription data from Whisper. The results show significant improvements across most entity categories, especially for underrepresented classes like Law, Facility, and Product, with near-perfect scores for some high-frequency categories. This work demonstrates the value of domain-specific, manually annotated data for boosting NER performance in Catalan and suggests a scalable path for expanding Catalan NLP applications in media, governance, and culture.

Abstract

Named Entity Recognition (NER) is a critical component of Natural Language Processing (NLP) for extracting structured information from unstructured text. However, for low-resource languages like Catalan, the performance of NER systems often suffers due to the lack of high-quality annotated datasets. This paper introduces NERCat, a fine-tuned version of the GLiNER[1] model, designed to improve NER performance specifically for Catalan text. We used a dataset of manually annotated Catalan television transcriptions to train and fine-tune the model, focusing on domains such as politics, sports, and culture. The evaluation results show significant improvements in precision, recall, and F1-score, particularly for underrepresented named entity categories such as Law, Product, and Facility. This study demonstrates the effectiveness of domain-specific fine-tuning in low-resource languages and highlights the potential for enhancing Catalan NLP applications through manual annotation and high-quality datasets.