README.md

metadata

license: openrail
language:
  - ar
  - en
tags:
  - quran
  - islam
  - religion
  - arabic
size_categories:
  - 1M<n<10M

📖 Comprehensive Quran Dataset

Welcome to the Comprehensive Quran Dataset. This repository contains a highly structured, developer-friendly, and machine-learning-ready dataset of the Holy Quran, along with its translations.

🗂 Dataset Structure

The dataset is designed to serve both AI researchers and web/mobile app developers:

1. The Core Dataset (all_ayat.jsonl)

• Contains all Ayahs (verses) of the Quran in a flattened JSONL format.
• Optimized for Hugging Face's datasets library and AI training pipelines.
• Instantly viewable via the Hugging Face Dataset Viewer.

2. By Surah (by_sura/)

• Contains 114 JSON files, one for each Surah.
• Ideal for developers who need to fetch a specific Surah directly as a static API without downloading the entire dataset.

3. By Translation (by_translation/)

• Contains the meanings and translations of the Quran categorized by language and translator.

4. Metadata and Indexes

• manifest.json and translations_index.json provide a quick map of the dataset's contents.

🚀 How to Use (Python / AI)

Using the Hugging Face datasets library, you can load the entire JSONL dataset in one line:

from datasets import load_dataset

# Load the core dataset
dataset = load_dataset("Islamiccontent/Quran-Dataset")
print(dataset['train'][0])

💻 How to Use (Web & Mobile Developers)

You can fetch specific JSON files dynamically in your app using the Hugging Face raw file URL. For example, to fetch Surah Al-Fatiha (Surah 1):

const response = await fetch("https://huggingface.co/datasets/Islamiccontent/Quran-Dataset/resolve/main/by_sura/1.json");
const surah = await response.json();
console.log(surah);