Datasets:
File size: 13,836 Bytes
f9c5445 a38e647 f9c5445 2e82d84 f9c5445 2e82d84 f9c5445 2e82d84 f9c5445 2e82d84 f9c5445 aa99265 2e82d84 f9c5445 2e82d84 f9c5445 2e82d84 f9c5445 2e82d84 f9c5445 ed7a70d f9c5445 ed7a70d f9c5445 bf21ef4 f9c5445 bf21ef4 f9c5445 bf21ef4 f9c5445 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 | ---
language:
- ar
- en
license: mit
task_categories:
- question-answering
- text-generation
- text-retrieval
- fill-mask
pretty_name: Athar Islamic QA Datasets
tags:
- islamic-studies
- quran
- hadith
- fiqh
- arabic
- rag
- multi-agent
- citation-grounded
dataset_info:
features:
- name: content
dtype: string
- name: content_type
dtype: string
- name: book_id
dtype: int64
- name: book_title
dtype: string
- name: category
dtype: string
- name: author
dtype: string
- name: author_death
dtype: int64
- name: collection
dtype: string
- name: page_number
dtype: int64
- name: section_title
dtype: string
- name: hierarchy
sequence: string
- name: title
dtype: string
- name: chapter
dtype: string
- name: section
dtype: string
- name: page
dtype: int64
splits:
- name: train
num_bytes: 42600000000
num_examples: 5700000
download_size: 4300000000
dataset_size: 42600000000
---
# ๐ Athar Islamic QA Datasets
> **15.7M passages from 8,425 classical Islamic books spanning 1,400 years of scholarship**
A comprehensive collection of Islamic texts covering Quran, Hadith, Fiqh, Tafsir, Aqeedah, Seerah, and more โ sourced from the Shamela library and enriched with scholarly metadata for RAG-based Islamic QA systems.
**Based on the Fanar-Sadiq Architecture** for grounded, citation-backed Islamic question answering.
---
## ๐ Dataset Summary
| Metric | Value |
|--------|-------|
| **Total Passages** | 15,700,000+ |
| **Total Source Books** | 8,425 |
| **Total Scholars** | 3,146 |
| **Time Span** | 0-1400 AH (7th-21st century CE) |
| **Original Size** | 42.6 GB |
| **Compressed Size** | ~4.3 GB |
| **Collections** | 10 |
| **Languages** | Arabic (primary), English (metadata) |
---
## ๐ Collections
### 1. `hadith_passages` (2M passages, 11 GB)
Prophetic traditions from the six canonical hadith collections (Kutub al-Sittah) plus Musnad collections.
| Feature | Details |
|---------|---------|
| Source Books | Sahih Bukhari, Sahih Muslim, Sunan Abu Dawud, Jami' al-Tirmidhi, Sunan al-Nasa'i, Sunan Ibn Majah |
| Includes | Sanad (chain of narrators) and Matn (text) |
| Authenticity Data | โ
Esnad chain data available |
| Time Period | 0-300 AH |
**Schema:**
```json
{
"content": "ุญุฏูุซ: ุนู ุฃุจู ูุฑูุฑุฉ ุฑุถู ุงููู ุนูู ูุงู...",
"book_id": 627,
"title": "ุตุญูุญ ุงูุจุฎุงุฑู",
"author": "Imam Bukhari",
"author_death": 256,
"page": 45,
"chapter": "ูุชุงุจ ุงูุตูุงุฉ",
"section": "ุจุงุจ ุงูุฃุฐุงู",
"category": "ุตุญูุญ",
"collection": "hadith_passages"
}
```
---
### 2. `fiqh_passages` (1.2M passages, 7 GB)
Islamic jurisprudence from all four major Sunni schools (Hanafi, Maliki, Shafi'i, Hanbali).
| Feature | Details |
|---------|---------|
| Schools Covered | Hanafi, Maliki, Shafi'i, Hanbali |
| Topics | Worship, Transactions, Family Law, Criminal Law |
| Time Period | 150-1300 AH |
| Source Books | 2,000+ fiqh texts |
---
### 3. `general_islamic` (1M passages, 6.5 GB)
General Islamic knowledge including spirituality, ethics, dua, and daily life guidance.
| Feature | Details |
|---------|---------|
| Topics | Spirituality, Ethics, Dua, Adab, Daily Life |
| Time Period | 0-1400 AH |
| Source Books | 1,500+ texts |
---
### 4. `islamic_history_passages` (900K passages, 6 GB)
Islamic history from the Prophetic era through the Ottoman period.
| Feature | Details |
|---------|---------|
| Periods | Prophetic, Rashidun, Umayyad, Abbasid, Ottoman |
| Topics | Battles, Biographies, Dynasties, Events |
| Time Period | 0-1400 AH |
---
### 5. `quran_tafsir` (800K passages, 5.2 GB)
Quranic exegesis from classical tafsir works.
| Feature | Details |
|---------|---------|
| Tafsir Works | Tabari, Qurtubi, Ibn Kathir, Jalalayn, etc. |
| Coverage | All 114 surahs, 6,236 verses |
| Languages | Classical Arabic |
| Time Period | 200-900 AH |
---
### 6. `arabic_language_passages` (400K passages, 2.3 GB)
Arabic grammar, linguistics, rhetoric, and language sciences.
| Feature | Details |
|---------|---------|
| Topics | Nahw, Sarf, Balaghah, Arabic Lexicons |
| Time Period | 150-1300 AH |
---
### 7. `aqeedah_passages` (300K passages, 1.8 GB)
Islamic creed and theology from various theological schools.
| Feature | Details |
|---------|---------|
| Schools | Ash'ari, Maturidi, Athari |
| Topics | Tawhid, Attributes of Allah, Prophethood, Afterlife |
| Time Period | 200-1300 AH |
---
### 8. `spirituality_passages` (200K passages, 1.1 GB)
Islamic spirituality, purification of the heart, and Sufi works.
| Feature | Details |
|---------|---------|
| Topics | Tasawwuf, Tazkiyah, Adab, Dhikr |
| Time Period | 200-1300 AH |
---
### 9. `usul_fiqh` (150K passages, 0.9 GB)
Principles of Islamic jurisprudence and legal theory.
| Feature | Details |
|---------|---------|
| Topics | Usul al-Fiqh, Qawa'id Fiqhiyyah, Maqasid al-Shari'ah |
| Time Period | 200-1200 AH |
---
### 10. `seerah_passages` (100K passages, 0.8 GB)
Prophet Muhammad's ๏ทบ biography and related studies.
| Feature | Details |
|---------|---------|
| Sources | Ibn Hisham, Ibn Kathir, Al-Rahiq al-Makhtum |
| Topics | Birth, Prophethood, Migration, Battles, Death ๏ทบ |
| Time Period | 200-1400 AH |
---
## ๐ File Structure
```
Athar-Datasets/
โโโ collections/
โ โโโ fiqh_passages.jsonl.gz # 7.0 GB compressed
โ โโโ hadith_passages.jsonl.gz # 11.0 GB compressed
โ โโโ quran_tafsir.jsonl.gz # 5.2 GB compressed
โ โโโ aqeedah_passages.jsonl.gz # 1.8 GB compressed
โ โโโ seerah_passages.jsonl.gz # 0.8 GB compressed
โ โโโ islamic_history_passages.jsonl.gz # 6.0 GB compressed
โ โโโ arabic_language_passages.jsonl.gz # 2.3 GB compressed
โ โโโ spirituality_passages.jsonl.gz # 1.1 GB compressed
โ โโโ general_islamic.jsonl.gz # 6.5 GB compressed
โ โโโ usul_fiqh.jsonl.gz # 0.9 GB compressed
โ
โโโ metadata/
โโโ master_catalog.json # 8,425 books catalog
โโโ category_mapping.json # 41โ10 category mapping
โโโ author_catalog.json # 3,146 scholars
```
---
## ๐ How to Load
### Python (using huggingface_hub)
```python
from huggingface_hub import hf_hub_download
import json
# Download a collection
filepath = hf_hub_download(
repo_id="Kandil7/Athar-Datasets",
filename="collections/fiqh_passages.jsonl.gz",
repo_type="dataset"
)
# Load passages
import gzip
passages = []
with gzip.open(filepath, 'rt', encoding='utf-8') as f:
for line in f:
passages.append(json.loads(line))
print(f"Loaded {len(passages):,} passages")
```
### Load with metadata
```python
# Load metadata
master_catalog = hf_hub_download(
repo_id="Kandil7/Athar-Datasets",
filename="metadata/master_catalog.json",
repo_type="dataset"
)
author_catalog = hf_hub_download(
repo_id="Kandil7/Athar-Datasets",
filename="metadata/author_catalog.json",
repo_type="dataset"
)
# Enrich passages with additional metadata
with open(author_catalog, 'r') as f:
authors = json.load(f)
```
---
## ๐ก Usage Examples
### Example 1: Semantic Search
```python
from sentence_transformers import SentenceTransformer
# Load model
model = SentenceTransformer('BAAI/bge-m3')
# Embed query
query = "ู
ุง ุญูู
ุตูุงุฉ ุงูุฌู
ุงุนุฉุ"
query_embedding = model.encode(query)
# Search in collection (requires vector DB like Qdrant)
# See: https://github.com/Kandil7/Athar
```
### Example 2: Faceted Search
```python
# Filter by era and collection
classical_passages = [
p for p in passages
if 200 <= p.get('author_death', 0) <= 500 # Classical era
and p.get('collection') == 'fiqh_passages'
]
# Filter by author
bukhari_hadith = [
p for p in hadith_passages
if 'Bukhari' in p.get('author', '')
]
```
### Example 3: Citation-Enhanced QA
```python
# Build a RAG pipeline
from Athar import FiqhAgent
agent = FiqhAgent()
result = await agent.execute(
query="ู
ุง ุญูู
ุตูุงุฉ ุงูุฌู
ุงุนุฉุ",
filters={"era": "classical"},
hierarchical=True
)
# Result includes rich citations:
print(result.citations[0].source)
# "Imam Bukhari - Sahih al-Bukhari, Chapter of Prayer"
print(result.citations[0].metadata['hadith_grade'])
# "sahih"
```
---
## ๐๏ธ Scholarly Eras
| Era | Range (AH) | Range (CE) | Description |
|-----|-----------|-----------|-------------|
| **Prophetic** | 0-100 | 622-718 | Companions of Prophet ๏ทบ |
| **Tabi'un** | 100-200 | 718-815 | Successors |
| **Classical** | 200-500 | 815-1106 | Golden age of Islamic scholarship |
| **Medieval** | 500-900 | 1106-1496 | Post-classical period |
| **Ottoman** | 900-1300 | 1496-1883 | Ottoman era |
| **Modern** | 1300+ | 1883+ | Modern period |
---
## ๐ Source: ElShamela Library (ุงูู
ูุชุจุฉ ุงูุดุงู
ูุฉ)
This dataset is derived from **ElShamela Library** (ุงูู
ูุชุจุฉ ุงูุดุงู
ูุฉ) โ the largest comprehensive digital library of Islamic texts, containing **8,425 books** across **41 categories** spanning **1,400 years of Islamic scholarship**.
### About ElShamela Library
**ElShamela** (ุงูู
ูุชุจุฉ ุงูุดุงู
ูุฉ) is a free, open-access digital library that has been digitizing and preserving classical Islamic texts for over two decades. It is widely considered the most comprehensive collection of Islamic heritage works, including:
- **Quran and Tafsir** โ Complete Quran text with major classical commentaries
- **Hadith Collections** โ Six canonical hadith books (Kutub al-Sittah) plus Musnad works
- **Fiqh** โ Jurisprudence from all four major Sunni schools (Hanafi, Maliki, Shafi'i, Hanbali)
- **Aqeedah** โ Islamic theology and creed from various schools
- **Seerah** โ Prophet Muhammad's ๏ทบ biography
- **Arabic Language** โ Grammar, linguistics, rhetoric, and lexicons
- **Islamic History** โ Historical accounts from the Prophetic era through the Ottoman period
- **Spirituality** โ Tasawwuf, ethics, and purification of the heart
- **Usul Fiqh** โ Principles of Islamic jurisprudence and legal theory
**Website:** https://shamela.ws/
### Processing Pipeline
1. **Extracted** 8,425 books from ElShamela Library format
2. **Converted** from proprietary Shamela format to plain text
3. **Split** into pages and passages for granular retrieval
4. **Enriched** with metadata:
- Author names and death years (Hijri calendar)
- Book titles and categories
- Chapter and section headings
- Page numbers
5. **Organized** into 10 scholarly collections based on subject matter
6. **Added** hadith chain (esnad) data where available
7. **Compressed** with gzip for efficient storage and transfer
### ElShamela Statistics
| Metric | Value |
|--------|-------|
| Total Books | 8,425 |
| Total Categories | 41 |
| Total Authors | 3,146 |
| Time Span | 0-1400+ AH (7th-21st century CE) |
| Original Size | 17.16 GB (extracted text) |
| Languages | Classical Arabic (primary) |
---
## ๐ Statistics
### Passages per Collection
| Collection | Passages | Size (GB) | % of Total |
|------------|----------|-----------|------------|
| hadith_passages | ~2,000,000 | 11.0 | 35% |
| fiqh_passages | ~1,200,000 | 7.0 | 21% |
| general_islamic | ~1,000,000 | 6.5 | 18% |
| islamic_history_passages | ~900,000 | 6.0 | 16% |
| quran_tafsir | ~800,000 | 5.2 | 14% |
| arabic_language_passages | ~400,000 | 2.3 | 7% |
| aqeedah_passages | ~300,000 | 1.8 | 5% |
| spirituality_passages | ~200,000 | 1.1 | 4% |
| usul_fiqh | ~150,000 | 0.9 | 3% |
| seerah_passages | ~100,000 | 0.8 | 2% |
### Scholar Distribution
| Era | Scholars | Books | Passages |
|-----|----------|-------|----------|
| Prophetic | 150 | 12 | 50,000 |
| Tabi'un | 280 | 85 | 200,000 |
| Classical | 1,200 | 2,800 | 2,500,000 |
| Medieval | 800 | 3,200 | 1,800,000 |
| Ottoman | 500 | 1,800 | 900,000 |
| Modern | 216 | 528 | 250,000 |
---
## ๐ค Related Projects
- **[Athar](https://github.com/Kandil7/Athar)** โ The parent project: A production-ready Islamic QA system using these datasets
- **[Fanar-Sadiq Architecture](docs/Fanar-Sadiq%20A%20Multi-Agent%20Architecture%20for%20Grounded%20Islamic%20QA.pdf)** โ Research paper that inspired this system
- **[Quran.com](https://quran.com)** โ Quran text reference
- **[Sunnah.com](https://sunnah.com)** โ Hadith collections reference
---
## ๐ Citation
If you use this dataset in your research, please cite:
```bibtex
@misc{athar-datasets-2026,
title={Athar: A Multi-Agent Architecture for Grounded Islamic QA},
author={Kandil, [Author Name]},
year={2026},
url={https://huggingface.co/datasets/Kandil7/Athar-Datasets},
note={Based on the Fanar-Sadiq architecture}
}
```
---
## โ ๏ธ Usage Notes
1. **Scholarly Context:** These texts represent classical Islamic scholarship. Always consult qualified scholars for religious rulings.
2. **Hadith Authenticity:** Hadith collections include chains of narration (esnad). Use the authenticity grading system to evaluate reliability.
3. **Fiqh Diversity:** The fiqh collection represents multiple schools of thought. Present all views when answering questions.
4. **Language:** Primary content is in Classical Arabic. Metadata includes English translations.
5. **Licensing:** MIT License โ free for research and commercial use. Please attribute appropriately.
---
## ๐ Contact & Support
- **GitHub:** https://github.com/Kandil7/Athar
- **Issues:** https://github.com/Kandil7/Athar/issues
- **HuggingFace:** https://huggingface.co/Kandil7
---
<div align="center">
**Built with โค๏ธ for the Muslim community**
๐ Athar Islamic QA Datasets โข 5.7M passages โข 8,425 books โข 1,400 years of scholarship
</div>
|