TajikNLPWorld/tajik-web-corpus

Dataset Card for Tajik Web Corpus

Dataset Details

Dataset Description

The Tajik Web Corpus is a large-scale collection of 319,298 documents in the Tajik language, totaling approximately 1.11 billion characters and 168.5 million words. The data has been cleaned, normalized, and deduplicated, and is provided in JSONL format with the following fields: title, text, category, source, date, and URL. It covers various domains including news, Wikipedia, government sources, social media, and literature, making it suitable for a wide range of NLP tasks.

• Curated by: Arabov Mullosharaf Kurbonovich
• Language(s) (NLP): Tajik (tg), with some Persian (fa) content
• License: other – see Licensing & Legal Notice below.

Licensing & Legal Notice

This dataset follows the practice established by large web-crawled corpora such as HPLT and OSCAR:

• Original source texts (documents, articles, etc.) remain the property of their respective authors and publishers. They are not owned by the dataset curator and are not covered by the CC BY‑SA 4.0 license.
• The structured compilation, metadata, and any original annotations created during dataset preparation are released under the CC BY‑SA 4.0 license.
• Users are solely responsible for ensuring their use of the underlying texts complies with applicable copyright law. For commercial use of verbatim excerpts, permission from the original copyright holders may be required.
• A notice‑and‑takedown procedure is in place: rights holders can request removal of specific content by contacting the dataset maintainers (see Dataset Card Contact). We commit to responding within 14 business days and removing disputed content in the next release.

Dataset Sources

• Repository: https://huggingface.co/datasets/TajikNLPWorld/tajik-web-corpus
• Papers:
  • Arabov, M. K. (2026). Benchmarking Parameter-Efficient Fine-Tuning of Large Language Models for Low-Resource Tajik Text Generation with the Tajik Web Corpus. arXiv preprint arXiv:2605.03742.
  • Arabov, M. K., Pirov, S. S., & Sultonov, B. (2026). Large Language Models for Low-Resource Languages: A Conceptual Framework for an Electronic Explanatory Dictionary of the Tajik Language. arXiv preprint arXiv:2608.04186.
  • Arabov, M. K., Habibullozoda, K., & Shirinov, N. (2026). TajikNLP: An Open-Source Toolkit for Comprehensive Text Processing of Tajik (Cyrillic Script). arXiv preprint arXiv:2605.04583.
  • Arabov, M. K., & Khaibullina, S. S. (2026). Analysis of the effectiveness of subword tokenizers in a low-resource linguistic environment: Implementation experience for the Tajik language. Russian Digital Libraries Journal, 29(2), 546–564.

Uses

Direct Use

The dataset is intended for training and evaluating NLP models for the Tajik language, including:

• Language modeling (pre‑training Transformer models like BERT, GPT)
• Text classification (topic, genre, sentiment)
• Machine translation (Tajik ↔ Persian, Tajik ↔ Russian)
• Information retrieval and text mining
• Linguistic studies of the Tajik language
• Named entity recognition and other token‑level tasks

Out-of-Scope Use

The dataset should not be used for:

• Identifying individuals or extracting personal information
• Generating harmful or misleading content
• Tasks requiring fine‑grained temporal information (only 0.3% of documents have dates)
• Any purpose that violates the original copyright of document texts (see Licensing & Legal Notice)

Dataset Structure

Data Fields

Field	Type	Coverage	Description
title	string	99.7%	Document title
text	string	100%	Full document text (includes title when available)
category	string	50.9%	Thematic category (e.g., Сиёсат, Иқтисод, Ҷамъият)
source	string	100%	Original source file name
date	string (YYYY-MM-DD)	0.3%	Publication date (mainly from books corpus)
url	string	100%	Original URL

Data Splits

Split	Documents	Size (JSONL)	Description
full	319,298	1.93 GB	Complete corpus
sample	1,000	6.1 MB	Random sample for testing/development

Dataset Creation

Curation Rationale

The corpus was created to address the lack of large‑scale, clean, and publicly available text data for the Tajik language. By aggregating diverse sources and applying extensive cleaning and normalization, the dataset provides a solid foundation for NLP research in Tajik.

Source Data

Data Collection and Processing

Data was collected from publicly accessible websites and archives. The processing pipeline included:

1. Scraping / Collection from various sources (Wikipedia, news sites, social media, etc.)
2. Initial Cleaning (removal of HTML tags, URLs, emails, non‑textual elements)
3. Character Normalization to standard Tajik Cyrillic
4. Quality Filtering (minimum length, language filtering)
5. Title Cleaning (removal of garbage titles)
6. Category Normalization
7. Final Assembly into JSONL

Who are the source data producers?

The original texts were produced by:

• Wikipedia – community contributors
• News portals – journalists and media outlets (Asia-Plus, Khovar, Ozodi, etc.)
• Government sites – official press releases and documents
• Social media – public posts (Telegram, etc.)
• Books and literature – digitized works

Annotations

No additional manual annotations were added; the category field was derived from the source metadata where available.

Personal and Sensitive Information

Efforts were made to exclude personally identifiable information. However, news articles may mention names of public figures. Users should be aware of this.

Bias, Risks, and Limitations

• Genre bias: Dominated by news (80%) and Wikipedia (25%), with less social media and literature.
• Geographic bias: Content mostly from Tajikistan; other Tajik‑speaking regions underrepresented.
• Temporal bias: Sparse date information limits time‑based analysis.
• Category imbalance: Some categories overrepresented.
• Copyright constraints: The underlying texts may be protected; see Licensing & Legal Notice.

Recommendations

• For balanced training, consider subsampling or combining with other corpora.
• Be cautious with categories; about 49% of documents lack them.
• Do not rely on dates for temporal tasks.
• Respect original copyrights when redistributing excerpts.

Citation

BibTeX (dataset):

@misc{tajikwebcorpus2026,
  author = {Arabov, Mullosharaf Kurbonovich},
  title = {Tajik Web Corpus},
  year = {2026},
  publisher = {Hugging Face},
  url = {https://huggingface.co/datasets/TajikNLPWorld/tajik-web-corpus}
}

APA (dataset): Arabov, M. K. (2026). Tajik Web Corpus [Data set]. Hugging Face. https://huggingface.co/datasets/TajikNLPWorld/tajik-web-corpus

Related Publications

BibTeX:

@article{arabov2026benchmarking,
  author = {Arabov, Mullosharaf K.},
  title = {Benchmarking Parameter-Efficient Fine-Tuning of Large Language Models for Low-Resource Tajik Text Generation with the Tajik Web Corpus},
  journal = {arXiv preprint arXiv:2605.03742},
  year = {2026}
}

@article{arabov2026llm,
  author = {Arabov, Mullosharaf K. and Pirov, S. S. and Sultonov, B.},
  title = {Large Language Models for Low-Resource Languages: A Conceptual Framework for an Electronic Explanatory Dictionary of the Tajik Language},
  journal = {arXiv preprint arXiv:2608.04186},
  year = {2026}
}

@article{arabov2026tajiknlp,
  author = {Arabov, Mullosharif K. and Habibullozoda, Karomatullo and Shirinov, Nurali},
  title = {TajikNLP: An Open-Source Toolkit for Comprehensive Text Processing of Tajik (Cyrillic Script)},
  journal = {arXiv preprint arXiv:2605.04583},
  year = {2026}
}

@article{arabov2026tokenizers,
  author = {Arabov, M. K. and Khaibullina, S. S.},
  title = {Analysis of the effectiveness of subword tokenizers in a low-resource linguistic environment: Implementation experience for the Tajik language},
  journal = {Russian Digital Libraries Journal},
  volume = {29},
  number = {2},
  pages = {546--564},
  year = {2026}
}

APA:

• Arabov, M. K. (2026). Benchmarking Parameter-Efficient Fine-Tuning of Large Language Models for Low-Resource Tajik Text Generation with the Tajik Web Corpus. arXiv preprint arXiv:2605.03742.
• Arabov, M. K., Pirov, S. S., & Sultonov, B. (2026). Large Language Models for Low-Resource Languages: A Conceptual Framework for an Electronic Explanatory Dictionary of the Tajik Language. arXiv preprint arXiv:2608.04186.
• Arabov, M. K., Habibullozoda, K., & Shirinov, N. (2026). TajikNLP: An Open-Source Toolkit for Comprehensive Text Processing of Tajik (Cyrillic Script). arXiv preprint arXiv:2605.04583.
• Arabov, M. K., & Khaibullina, S. S. (2026). Analysis of the effectiveness of subword tokenizers in a low-resource linguistic environment: Implementation experience for the Tajik language. Russian Digital Libraries Journal, 29(2), 546–564.

Glossary

• Tajik language – a variety of Persian spoken primarily in Tajikistan, written in Cyrillic script.
• Low-resource language – a language with limited digital resources and tools.

More Information

For questions, contributions, or feedback, please open an issue on the Hugging Face repository.

Dataset Card Authors

• Arabov Mullosharaf Kurbonovich (TajikNLPWorld)

Dataset Card Contact

For questions, takedown requests, or collaboration, please open an issue on the Hugging Face repository:
https://huggingface.co/datasets/TajikNLPWorld/tajik-web-corpus/discussions
📧 Direct email: cool.araby@gmail.com