davidfadah/Hausa-NLP-Annotation

Hausa language data annotation samples and guidelines for Natural Language Processing (NLP) tasks. # Hausa Linguistic Annotation for NLP A demonstration of high-accuracy data labeling for Natural Language Processing (NLP) tasks in the Hausa language. ## Project Overview This repository showcases standard annotation practices for low-resource languages. It focuses on the morphological and syntactical nuances of Hausa, providing a "gold standard" sample of 15 sentences annotated for: - **Tense-Aspect-Mood (TAM) Markers** (e.g., *ya, ta, za*) - **Verb Grade Nuances** (e.g., Grade 1 vs Grade 4) - **Idiomatic Semantic Mapping** ## Repository Structure - `hausa_corpus.json`: The core dataset formatted for machine learning models. - `annotation_guidelines.md`: Strict linguistic rules used to ensure consistency across the dataset. - `LICENSE`: Open-source Creative Commons (CC-BY-4.0) license. ## Technical Details - **Language**: Hausa (Standard) - **Format**: JSON (Structured for Token-level analysis) - **Tags**: TAM (Tense-Aspect-Mood), N (Noun), V (Verb), IDM (Idiom)