Is this dataset free to use?

Yes, it is hosted publicly on the Hugging Face Hub and can be loaded at no cost with the datasets library.

How do I access the data?

Use the Hugging Face datasets library with load_dataset('ieasybooks-org/prophet-mosque-library').

License information is not specified in the dataset card; check the repository for usage terms before commercial application.

prophet-mosque-library

Processed Islamic books from Prophet's Mosque Library in PDF, TXT, and DOCX.

DatasetText & NLP↓ 247K/moFree

Open dataset

Updated 2026-06-18

What is prophet-mosque-library?

This dataset consists of PDF files from the Prophet's Mosque Library along with TXT and DOCX versions generated via document AI processing.

It supports image-to-text and NLP research involving large-scale Arabic Islamic texts across religious and scholarly topics.

What you can build with prophet-mosque-library

Train domain-specific NLP models

Fine-tune transformers or embeddings on the extracted TXT files for tasks like classical Arabic understanding or Islamic topic classification.

Build semantic search over Islamic texts

Index the DOCX/TXT content to create retrieval systems that surface passages across 70+ categories of religious literature.

Perform large-scale text analytics

Run corpus-wide statistics, theme extraction, or cross-category comparisons using the 23 million pages of processed Islamic books.

Load prophet-mosque-library

Python

from datasets import load_dataset

ds = load_dataset("ieasybooks-org/prophet-mosque-library")

1pip install datasets
2from datasets import load_dataset
3ds = load_dataset('ieasybooks-org/prophet-mosque-library')
4Access 'train' split and load TXT or DOCX fields per example
5Iterate over subsets by category for targeted processing