This dataset is a cleaned Nande–French parallel corpus containing 14,478 aligned sentence pairs for machine translation research and development.

The corpus covers Nande (nnb) and French (fr) and is provided with a 90:10 train/test split:

Training set: 90%
Test set: 10%
Total: 14,478 aligned sentence pairs

Nande is a Bantu language spoken primarily in the Democratic Republic of the Congo (DRC). NLP resources for Nande remain relatively limited, and this corpus is intended to contribute to the development and evaluation of machine translation systems for the language.

Dataset Summary

Property	Value
Languages	Nande (`nnb`), French (`fr`)
Language family	Bantu
Task	Machine Translation
Dataset type	Parallel Corpus
Number of sentence pairs	14,478
Train/Test split	90% / 10%
License	CC BY 4.0
Source	Mozilla Data Collective
Translation contributors	CLEAR Global translator community

Dataset Structure

Each example consists of an aligned sentence pair:

Field	Language	Description
`nnb`	Nande	Sentence in Nande
`fr`	French	Corresponding French translation

The corpus contains 14,478 aligned Nande–French sentence pairs in total.

Source and Provenance

This dataset is a cleaned and prepared version of the corpus originally published through the Mozilla Data Collective.

According to the original dataset description:

Translations were performed by professionals and volunteers of CLEAR Global's (formerly Translators without Borders) translator community.

The present release is not a newly translated corpus. It is a cleaned and processed derivative of the original resource, prepared for use in machine translation experiments.

The original source should be consulted for additional information regarding the collection process, translators, and provenance of the data.

Data Cleaning and Preparation

The original corpus was cleaned and prepared to make it more suitable for machine translation experiments.

This release may therefore differ from the original dataset in terms of formatting, sentence alignment, filtering, or other preprocessing operations.

The resulting dataset contains 14,478 aligned sentence pairs.

Because this is a derived version of the original resource, users who require the unprocessed source data should refer to the original Mozilla Data Collective dataset.

Intended Use

This dataset is intended for:

Nande–French machine translation
French–Nande machine translation
Fine-tuning neural machine translation models
Evaluating machine translation systems
Low-resource language NLP research
Multilingual and cross-lingual NLP research
Development of language technologies for African languages
Research on Nande language processing

The corpus was primarily used to fine-tune the pretrained T5-base model for the development of the following translation model:

SalomonMetre13/nnb_fr_mt_v3

Limitations

Nande is a relatively low-resource language in natural language processing, and the size of this corpus limits the amount of linguistic and domain variation available for training machine translation systems.

The dataset may not fully represent:

The diversity of Nande linguistic usage
Regional or dialectal variation
Different registers and writing styles
Specialized domains
Informal or conversational language
The full range of Nande-speaking communities

Models trained on this corpus may therefore have limited generalization to domains or linguistic contexts that are not represented in the data.

Translation quality should be evaluated carefully, especially for specialized, culturally sensitive, or high-stakes applications.

The dataset should not be considered a comprehensive representation of the Nande language or of all Nande-speaking communities.

Bias and Representation

The composition of a parallel corpus is influenced by the sources from which its sentences were collected and translated. Consequently, the linguistic patterns, vocabulary, topics, and registers represented in this dataset may not reflect the full diversity of Nande or French.

Users should take these potential representation limitations into account when training or evaluating models with this dataset.

Licensing

This dataset is released under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.

Because this release is derived from an existing dataset, users should also review the licensing and attribution requirements of the original Mozilla Data Collective resource.

For more information about the original dataset, see:

Mozilla Data Collective — Nande–French Corpus

Acknowledgements

We acknowledge the contributors to the original corpus, including the professional translators and volunteers of CLEAR Global's translator community (formerly Translators without Borders), whose translation work forms the basis of this resource.

We also acknowledge the Mozilla Data Collective for hosting and distributing the original dataset.