Datasets:

Modalities:
Tabular
Text
Formats:
parquet
Languages:
Igbo
DOI:
Libraries:
Datasets
pandas
License:
igbo / README.md
0xnu's picture
Update README.md
55e5401 verified
metadata
language:
  - ig
tags:
  - language
  - igbo
size_categories:
  - 500K<n<2M
license: apache-2.0

Igbo Dataset

The vocabulary foundation is organized by linguistic categories (pronouns, verbs, nouns, adjectives) with over 200 core Igbo words.

Linguistic Accuracy: The vocabulary accurately reflects Igbo grammar patterns, including:

  • Proper tone marks (ị, ọ, ụ, ṅ)
  • Igbo-specific grammatical constructions
  • Present continuous tense with "na-" prefix
  • Appropriate verb conjugations and noun formations

Cultural Integration: Unlike generic datasets, this includes:

  • Traditional Igbo proverbs (ilu) with explanations
  • Cultural practices like Iri ji (New Yam Festival)
  • Marriage customs and bride price traditions
  • Traditional greetings and social interactions

Data Types

  • Synthetic sentences (35%) - Basic vocabulary combinations
  • Pattern variations (18%) - Grammar pattern practice
  • Conversations (15%) - Real dialogue examples
  • Q&A pairs (12%) - Knowledge testing
  • Translations (8%) - Igbo-English pairs
  • Grammar (5%) - Linguistic structure
  • Culture (4%) - Traditional knowledge
  • Stories (2%) - Narrative completion
  • Proverbs (1%) - Wisdom sayings

Citation

If you find this work relevant or helpful to your work, please kindly cite it:

@misc{igbodataset,
  title={Igbo Dataset}, 
  author={Finbarrs Oketunji},
  year={2025}
}

Copyright

(c) Copyright 2025 Finbarrs Oketunji. All Rights Reserved.