---
license: mit
language:
- si
tags:
- sinhala
- wikipedia
---

# Sinhala Wikipedia Article Titles Dataset

This repository contains a text file (`sinhala_article_titles_cleaned.txt`) listing the titles of approximately 38,000 articles from the Sinhala Wikipedia (`si.wikipedia.org`). The dataset was created to support natural language processing (NLP) tasks, particularly for fine-tuning language models for the Sinhala language.

## Dataset Details
- **File**: `sinhala_article_titles.txt`
- **Format**: Plain text, one title per line, encoded in UTF-8.
- **Source**: Extracted from the Sinhala Wikipedia using the MediaWiki API.
- **Size**: Approximately 38,000 titles (subject to Wikipedia's current article count).
- **Purpose**: Intended for use in scraping article content or as a reference for Sinhala NLP research.

## Usage
The titles can be used to:
- Fetch full article content via the MediaWiki API or web scraping (ensure compliance with Wikipedia's terms).
- Generate a comprehensive dataset for training or fine-tuning language models.

Usage in Python:
```python
with open("sinhala_article_titles.txt", "r", encoding="utf-8") as f:
    titles = f.read().splitlines()
print(f"Number of titles: {len(titles)}")