NaijaNLP: A Survey of Nigerian Low-Resource Languages
TL;DR
This study presents the first comprehensive review of the state of affairs in NLP research across the three major Nigerian languages (NaijaNLP), quantitatively assess the available linguistic resources and identify key challenges.
Abstract
With over 500 languages in Nigeria, three languages - Hausa, Yor\`ub\'a and Igbo spoken by more than 175 million people, account for about 65% of the languages. However, these languages are classed as low-resource due to insufficient digital resources to support tasks in computational linguistics. While several research efforts and initiatives have been presented, a coherent understanding of the state of classic Natural Language Processing (NLP) spanning grammatical formalisation to linguistic resources that support models development is lacking. This study presents the first comprehensive review of the state of affairs in NLP research across the three major Nigerian languages (NaijaNLP). We quantitatively assess the available linguistic resources and identify key challenges. Of the 293 reviewed studies, 27.6% contributed new linguistic resources. This finding highlights a strong reliance on repurposing existing data rather than creating new resources. Also, language-specific challenges, such as morphological analysis and effective representation of diacritics, remain under-explored. To advance NaijaNLP and LR-NLP more broadly, we echo the need for more collaborative efforts in resource enrichment, comprehensive annotation, and increased community support.
