drelhaj/Arabic-news-and-management-corpus

Arabic Management, Economics & Financial News Corpus (1,200 Articles)

This corpus contains 1,200 Arabic news and management articles drawn from three distinct domains. It was originally compiled as part of research into Arabic Corpus Linguistics, management communication, financial discourse and domain-specific NLP. Both plain text and POS-tagged versions are available.

The dataset has been widely used in teaching and research, including the King Saud University book Corpus Linguistics: Analytical Applications on Arabic Language.

---

📘 Corpus Composition (1,200 Articles)

The corpus is divided into three equal sub-corpora:

1. Management Corpus – 400 Articles

Chairmen’s and CEOs’ statements from major Middle Eastern companies.

2. Economics News – 400 Articles

General economics and business reporting from leading Arabic newspapers.

3. Financial Market News – 400 Articles

Stock-market and financial reporting sourced from Investing.com.

Each file is one article. All texts have also been POS-tagged using the Stanford Arabic POS Tagger.

---

📂 Directory Structure

management/
*.txt (400 articles)

economicsNews/
*.txt (400 articles)

financialNews/
*.txt (400 articles)

management_pos/
economicsNews_pos/
financialNews_pos/

---

🧹 CSV Version (Cleaned)

The release includes a cleaned, machine-learning-ready CSV: full_corpus.csv

Columns:

column	description
id	filename (article ID)
domain	management / economics / financial
text	full article text with all line breaks removed

All articles have been normalised by:

• removing internal line breaks
• collapsing multiple spaces
• producing one clean string per article

Train/Dev/Test splits

Prepared as: train.csv (80%) dev.csv (10%) test.csv (10%)

Splits preserve domain balance.

---

🔍 Loading Examples

Load CSV splits

from datasets import load_dataset

ds = load_dataset("YOUR_REPO", "csv_splits")
print(ds["train"][0])

Load raw text sub-corpora

ds = load_dataset("YOUR_REPO", "full_text")
print(ds["train"][0]["text"])

---

📖 Citation

Please cite the following work:

El-Haj, M. et al. (2017). Corpus Linguistics: Analytical Applications on Arabic Language. In Arabic: لغويات المدونة الحاسوبية - تطبيقات تحليلة على العربية الطبيعية. King Saud University Press, Riyadh, Saudi Arabia. PDF: https://elhaj.uk/docs/arabic_cl.pdf

---

⚠ Licence

This dataset is released under CC-BY 4.0 for research and educational use. Source articles originate from publicly accessible media outlets.

🙏 Acknowledgements

Developed at UCREL, Lancaster University, in support of research on Arabic corpus linguistics, management communication, financial analysis and Arabic NLP.