gurgutan/sunnah_ar_en_dataset

Dataset Card for Dataset Name

Dataset Card for Hadiths 14 Books Collection

This dataset contains a comprehensive bilingual (Arabic-English) collection of hadiths from 14 major authenticated books of Islamic tradition. It includes over 50762 narrations with complete metadata, organized by book, chapter, and narrator. Each hadith is presented in both its original Arabic text and English translation, making it ideal for cross-lingual NLP tasks, Islamic question-answering systems, and religious chatbots. The structured format allows for easy retrieval, citation, and integration into AI applications focused on Islamic knowledge.

Hadiths 14 books collection. Books included:

• Sahih al-Bukhari صحيح البخاري
• Sahih Muslim صحيح مسلم
• Sunan Abi Dawud سنن أبي داود
• Jami` at-Tirmidhi جامع الترمذي
• Sunan an-Nasa'i سنن النسائي
• Sunan Ibn Majah سنن ابن ماجه
• Muwatta Malik موطأ مالك
• Musnad Ahmad مسند أحمد
• Sunan ad-Darimi سنن الدارمي
• Riyad as-Salihin رياض الصالحين
• Shamail al-Muhammadiyah الشمائل المحمدية
• Bulugh al-Maram بلوغ المرام
• Al-Adab Al-Mufrad الأدب المفرد
• Mishkat al-Masabih مشكاة المصابيح

Dataset Details

Dataset Description

Hadith are the transmitted narrations concerning the speech, actions, appearance, and approvals of the Messenger of Allah, the Prophet Muhammad (peace and blessings be upon him).

• Curated by: Slepovichev Ivan
• Language(s) (NLP): [English, Arabic]
• License: MIT

Dataset Sources [optional]

• Repository: hadith-database

Uses

This dataset is used for research purposes.

Dataset Structure

The dataset consists of hadith records from 14 books, with each record containing metadata and text in both English and Arabic. Each entry in the dataset follows this structure:

{
  "book_id": "Unique identifier for the book",
  "book_title_en": "Book title in English",
  "book_author_en": "Book author name in English",
  "book_title_ar": "Book title in Arabic",
  "book_author_ar": "Book author name in Arabic",
  "hadith_uid": "Unique identifier for the hadith",
  "hadith_book_id": "Book identifier the hadith belongs to",
  "hadith_chapter_id": "Chapter identifier within the book",
  "hadith_chapter_name_ar": "Chapter name in Arabic",
  "hadith_chapter_name_en": "Chapter name in English",
  "hadith_text_ar": "Full hadith text in Arabic",
  "hadith_text_en": "Full hadith text in English",
  "hadith_narrator_en": "Name of the narrator in English"
}

Fields Description:

• book_id: Unique identifier for each of the 14 hadith books in the collection
• book_title_en/ar: Title of the hadith book in English and Arabic
• book_author_en/ar: Author of the hadith book in English and Arabic
• hadith_uid: Unique identifier for each hadith across the entire collection
• hadith_book_id: Identifier for hadith within the book containing this hadith
• hadith_chapter_id: Identifier for the chapter within the book
• hadith_chapter_name_en/ar: Name of the chapter in English and Arabic
• hadith_text_en/ar: The full text of the hadith in English and Arabic
• hadith_narrator_en: The name of the narrator in English

The dataset contains hadiths from 14 major collections, providing a comprehensive resource for hadith studies, translation tasks, and question-answering applications in both English and Arabic.

Dataset Creation

Curation Rationale

This dataset was curated to provide a comprehensive, bilingual collection of authentic hadiths from 14 major books for use in AI applications, particularly:

1. Islamic question-answering systems
2. Religious chatbots and virtual assistants
3. Cross-lingual information retrieval systems
4. Educational tools for Islamic studies

The parallel English-Arabic structure makes it especially valuable for developing systems that can process and respond to religious queries in multiple languages.

Source Data

Data Collection and Processing

The hadiths were collected from reliable Islamic sources and digitized with careful attention to authenticity and accuracy. The processing workflow included:

1. Extraction of hadiths from original Arabic sources
2. Verification against authenticated collections
3. Alignment with established English translations
4. Structuring into a consistent JSON format
5. Assignment of unique identifiers for cross-referencing
6. Organization by book, chapter, and narrative chain

The dataset maintains the original chapter structure of each book, preserving the contextual organization of the hadiths as intended by their original compilers.

Who are the source data producers?

The original hadiths were compiled by renowned Islamic scholars including Imam Bukhari, Imam Muslim, and others who meticulously collected and verified these narrations during the formative period of Islamic scholarship. The digital compilation and structuring was performed by Slepovichev Ivan, drawing from established translations and digital Islamic libraries.

Intended Uses

This dataset is specifically designed for:

• Training AI models to answer questions about Islamic teachings based on authentic hadiths
• Developing conversational agents that can discuss Islamic topics with proper scriptural references
• Creating search and retrieval systems for Islamic knowledge
• Supporting comparative religious studies and research
• Enabling multilingual access to hadith literature

Researchers and developers can use this dataset to train models that provide evidence-based responses to religious queries, with the ability to cite specific hadiths as sources for the information provided.

Bias, Risks, and Limitations

Recommendations

Users should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.

Citation

If you find our work helpful, feel free to give us a cite. Dataset was created by Slepvochev Ivan. If you use this dataset, please cite:

@dataset{quran_tafsir, title = {QuranDataset: A Dataset of Quran Verses and Tafseer},
url = {https://escape-team.tech/},
author = {Slepovichev Ivan},
email = {gurgutan@yandex.ru},
month = {March},
year = {2025} }

Dataset Card Contact

gurgutan@yandex.ru hi@escape-team.tech