Mutashabihat Analysis Examples
Setup
from qurankit import Quran
quran = Quran.load()
engine = quran.mutashabihat
Exact Repeated Phrases
# Find all exact repeated phrases (min 2 words)
matches = engine.find_exact_phrases(min_length=2)
for match in matches[:20]:
print(f"'{match.text}' ({match.length} words) x {match.count}")
for occ in match.occurrences[:3]:
print(f" {occ.reference}")
print()
# With minimum occurrences
matches = engine.find_exact_phrases(min_length=3, min_occurrences=3)
# Limit phrase length
matches = engine.find_exact_phrases(min_length=2, max_length=8)
# Specific surahs only
matches = engine.find_exact_phrases(min_length=2, surahs=[1, 2, 3])
Near-Duplicate Phrases
# Find phrases with minor variations
matches = engine.find_near_duplicates(
min_length=3,
max_distance=2, # Max Levenshtein distance
min_occurrences=2
)
for match in matches[:10]:
print(f"Base: '{match.base_text}'")
for variant in match.variants:
print(f" dist={variant.distance}: '{variant.text}'")
for occ in variant.occurrences[:2]:
print(f" {occ.reference}")
Similar Openings/Endings
# Similar openings to Ayat al-Kursi
openings = engine.find_similar_openings(2, 255, max_distance=2)
for opening in openings:
print(f"{opening.reference}: '{opening.text}' (dist: {opening.distance})")
# Similar endings
endings = engine.find_similar_endings(2, 255, max_distance=2)
# Compare surah openings
comparisons = engine.compare_surah_openings([1, 2, 3, 4], max_words=5)
for comp in comparisons:
print(f"Surah {comp.surah}: '{comp.opening}'")
Recurring Expressions
# Expressions recurring across corpus
expressions = engine.find_recurring_expressions(
min_occurrences=3,
min_length=2,
max_length=8
)
for expr in expressions[:20]:
print(f"'{expr.text}' x{expr.count} surahs:{expr.surahs_covered}")
for occ in expr.occurrences[:3]:
print(f" {occ.reference}")
Shared Roots/Lemmas
# Ayahs sharing roots with 2:255
shared = engine.find_shared_roots(2, 255, min_shared=2)
for item in shared[:10]:
print(f"{item.reference}: {item.shared_roots} shared roots")
for root in item.roots:
print(f" {root.root}: {root.count}x")
# Shared lemmas
shared = engine.find_shared_lemmas(2, 255, min_shared=2)
Word Order Variations
# Same words, different order
variations = engine.find_word_order_variations(
reference=(2, 255),
window=5
)
for var in variations:
print(f"{var.reference}: '{var.text}' (sim: {var.similarity:.2f})")
Comparison Report
from qurankit.analysis.reports import ComparisonReport
# Generate comprehensive report for Ayat al-Kursi
report = ComparisonReport.generate(
quran,
reference=(2, 255),
include_mutashabihat=True,
include_shared_roots=True,
include_shared_lemmas=True,
include_near_duplicates=True
)
# Export formats
print(report.to_markdown())
report.to_json("ayat_kursi_report.json")
report.to_csv("ayat_kursi_report.csv")
# Summary
print(f"Exact phrases: {report.exact_phrase_count}")
print(f"Near duplicates: {report.near_duplicate_count}")
print(f"Shared roots: {report.shared_root_count}")
print(f"Shared lemmas: {report.shared_lemma_count}")
Corpus Statistics
from qurankit.analysis.reports import CorpusStatistics
stats = CorpusStatistics.compute(quran)
print(f"Total surahs: {stats.total_surahs}")
print(f"Total ayahs: {stats.total_ayahs}")
print(f"Total words: {stats.total_words}")
print(f"Unique roots: {stats.unique_roots}")
print(f"Unique lemmas: {stats.unique_lemmas}")
print(f"Repeated phrases: {stats.repeated_phrase_count}")
print(f"Most repeated: '{stats.most_repeated_phrase.text}' ({stats.most_repeated_phrase.count}x)")
# Per-surah stats
for s in stats.surah_statistics[:10]:
print(f"Surah {s.number}: {s.ayah_count} ayahs, {s.word_count} words, {s.unique_roots} roots")
CLI Commands
# Exact phrases
qurankit mutashabihat phrases --min-length 2 --min-occurrences 3
# Similar to ayah
qurankit mutashabihat similar 2:255 --type openings --max-distance 2
qurankit mutashabihat similar 2:255 --type endings --max-distance 2
# Recurring expressions
qurankit mutashabihat recurring --min-length 2 --min-occurrences 3
# Shared roots
qurankit mutashabihat shared-roots 2:255 --min-shared 2
# Full report
qurankit mutashabihat report 2:255 --output report.md
# Corpus stats
qurankit mutashabihat stats
Research Best Practices
# Document your methodology
metadata = {
"method": "exact_phrase_matching",
"min_length": 3,
"min_occurrences": 2,
"text_edition": quran.current_edition.id,
"algorithm_version": engine.version,
"normalization": "source_text", # or "broad", etc.
"date": "2024-01-15"
}
# Always verify results against source text
matches = engine.find_exact_phrases(min_length=3, min_occurrences=2)
# Manual verification sample
for match in matches[:5]:
for occ in match.occurrences:
ayah = quran.ayah(occ.surah_number, occ.ayah_number)
# Verify match.text appears in ayah.text_original
assert match.text in ayah.text_original
# Report with caveats
print("Note: Textual similarity does not imply semantic/theological equivalence.")
print("Results are deterministic computational matches requiring scholarly review.")