​

Advanced Analytical Queries

​

1. Multi-dimensional Thematic Analysis

​

Description

​

Query

// Find verses that address multiple themes
MATCH (v:Verse)
MATCH (v)-[:ADDRESSES_TOPIC]->(t:Topic)
WHERE t.name IN $theme_list
WITH v, collect(DISTINCT t.name) AS themes
WHERE size(themes) >= $min_themes

// Get chapter information
MATCH (c:Chapter)-[:CONTAINS]->(v)

// Return results with theme combinations
RETURN v.verse_key, v.text_uthmani,
       c.name_english AS chapter_name,
       c.revelation_place AS revelation_place,
       themes
ORDER BY size(themes) DESC, v.verse_key
LIMIT $limit

​

Parameters

• theme_list: List of themes to analyze (e.g., [“Mercy”, “Justice”, “Faith”, “Accountability”])
• min_themes: Minimum number of themes that must be present (e.g., 2)
• limit: Maximum number of verses to return (e.g., 30)

​

Expected Results

​

Variations

• Analyze theme co-occurrence: WITH themes, count(v) AS verse_count RETURN themes, verse_count ORDER BY verse_count DESC
• Compare across revelation periods: WITH c.revelation_place AS place, themes, count(v) AS verse_count RETURN place, themes, verse_count ORDER BY place, verse_count DESC
• Create theme correlation matrix: UNWIND themes AS theme1 UNWIND themes AS theme2 WHERE theme1 < theme2 WITH theme1, theme2, count(v) AS co_occurrences RETURN theme1, theme2, co_occurrences ORDER BY co_occurrences DESC

​

2. Chronological Theme Evolution Analysis

​

Description

​

Query

// Match verses addressing the topic
MATCH (t:Topic {name: $topic_name})<-[:ADDRESSES_TOPIC]-(v:Verse)<-[:CONTAINS]-(c:Chapter)
WHERE c.revelation_order IS NOT NULL

// Group by revelation period
WITH c.revelation_order AS rev_order,
     CASE
       WHEN c.revelation_order <= 30 THEN "Early Meccan"
       WHEN c.revelation_order <= 85 THEN "Late Meccan"
       ELSE "Medinan"
     END AS period,
     c.name_english AS chapter,
     count(v) AS verse_count,
     collect(v.verse_key) AS verses

// Calculate cumulative metrics
WITH period, rev_order, chapter, verse_count, verses
ORDER BY rev_order
WITH period, collect({rev_order: rev_order, chapter: chapter, verse_count: verse_count, verses: verses}) AS chapters
UNWIND chapters AS chapter_data

// Return detailed evolution data
RETURN period,
       chapter_data.rev_order AS revelation_order,
       chapter_data.chapter AS chapter_name,
       chapter_data.verse_count AS verse_count,
       chapter_data.verses[0..3] AS sample_verses
ORDER BY revelation_order

​

Parameters

• topic_name: Name of the topic to analyze (e.g., “Faith”)

​

Expected Results

​

Variations

• Calculate density metrics: WITH period, sum(chapter_data.verse_count) AS period_verses, count(chapter_data) AS period_chapters RETURN period, period_verses, period_chapters, period_verses * 1.0 / period_chapters AS verses_per_chapter
• Compare multiple topics: MATCH (t:Topic)<-[:ADDRESSES_TOPIC]-(v:Verse)<-[:CONTAINS]-(c:Chapter) WHERE t.name IN $topic_list AND c.revelation_order IS NOT NULL WITH t.name AS topic, c.revelation_order AS rev_order, count(v) AS verse_count ORDER BY rev_order, topic RETURN rev_order, collect({topic: topic, count: verse_count}) AS topic_counts
• Analyze linguistic changes: MATCH (t:Topic {name: $topic_name})<-[:ADDRESSES_TOPIC]-(v:Verse)-[:CONTAINS_WORD]->(w:Word)-[:HAS_ROOT]->(r:Root) WITH period, r.text_clean AS root, count(w) AS occurrences WHERE occurrences > 3 RETURN period, collect({root: root, count: occurrences}) AS root_usage ORDER BY period

​

3. Structural Pattern Analysis

​

Description

​

Query

// Get verse word counts
MATCH (c:Chapter)-[:CONTAINS]->(v:Verse)-[:CONTAINS_WORD]->(w:Word)
WITH c, v, count(w) AS word_count

// Calculate chapter statistics
WITH c, avg(word_count) AS avg_length,
     min(word_count) AS min_length,
     max(word_count) AS max_length,
     stDev(word_count) AS std_dev,
     percentileCont(word_count, 0.5) AS median_length,
     count(v) AS verse_count

// Return chapter statistics
RETURN c.chapter_number, c.name_english, c.revelation_place,
       avg_length, median_length, min_length, max_length, std_dev,
       verse_count
ORDER BY c.chapter_number

​

Parameters

​

Expected Results

​

Variations

• Analyze verse position patterns: WITH c, v.verse_number AS position, word_count ORDER BY c.chapter_number, position WITH c.chapter_number AS chapter, collect({position: position, length: word_count}) AS verse_lengths RETURN chapter, verse_lengths
• Find structural anomalies: WITH c, v, word_count, avg(word_count) OVER (PARTITION BY c.id) AS chapter_avg WHERE abs(word_count - chapter_avg) > 2 * stDev(word_count) OVER (PARTITION BY c.id) RETURN c.name_english, v.verse_key, word_count, chapter_avg, word_count - chapter_avg AS deviation
• Compare Meccan vs Medinan structure: WITH c.revelation_place AS place, avg(word_count) AS avg_length, stDev(word_count) AS std_dev RETURN place, avg_length, std_dev

​

4. Semantic Coherence Analysis

​

Description

​

Query

// Get verses in a specific chapter
MATCH (c:Chapter {chapter_number: $chapter_number})-[:CONTAINS]->(v:Verse)
WITH c, v
ORDER BY v.verse_number

// Collect verses with their embeddings
WITH c, collect({key: v.verse_key, number: v.verse_number, embedding: v.embedding}) AS verses

// Calculate similarity between adjacent verses
UNWIND range(0, size(verses)-2) AS i
WITH c, verses, i,
     verses[i].key AS verse1,
     verses[i].number AS number1,
     verses[i+1].key AS verse2,
     verses[i+1].number AS number2,
     vector_similarity(verses[i].embedding, verses[i+1].embedding) AS similarity

// Identify significant transitions (low similarity points)
WITH c, verse1, number1, verse2, number2, similarity,
     avg(similarity) OVER () AS avg_similarity,
     stDev(similarity) OVER () AS std_dev
WHERE similarity < avg_similarity - $transition_threshold * std_dev

// Return potential thematic transitions
RETURN verse1, verse2, similarity,
       avg_similarity, std_dev,
       (avg_similarity - similarity) / std_dev AS transition_strength
ORDER BY transition_strength DESC
LIMIT $limit

​

Parameters

• chapter_number: Chapter number to analyze (e.g., 2)
• transition_threshold: Number of standard deviations below average to identify transitions (e.g., 1.5)
• limit: Maximum number of transitions to return (e.g., 10)

​

Expected Results

​

Variations

• Identify coherent sections: WHERE similarity > avg_similarity + $coherence_threshold * std_dev WITH verse1, verse2, similarity ORDER BY verse1 WITH collect({start: verse1, end: verse2, similarity: similarity}) AS coherent_pairs
• Compare with topic transitions: MATCH (v1:Verse {verse_key: verse1})-[:ADDRESSES_TOPIC]->(t1:Topic) WITH verse1, verse2, similarity, transition_strength, collect(DISTINCT t1.name) AS topics1 MATCH (v2:Verse {verse_key: verse2})-[:ADDRESSES_TOPIC]->(t2:Topic) WITH verse1, verse2, similarity, transition_strength, topics1, collect(DISTINCT t2.name) AS topics2 RETURN verse1, verse2, similarity, transition_strength, topics1, topics2, size([x IN topics1 WHERE NOT x IN topics2]) AS topic_changes
• Analyze global coherence: WITH c.chapter_number AS chapter, avg(similarity) AS coherence, min(similarity) AS min_coherence, max(similarity) AS max_coherence RETURN chapter, coherence, min_coherence, max_coherence ORDER BY coherence DESC

​

5. Network Analysis of Thematic Relationships

​

Description

​

Query

// Calculate how many verses address each topic
MATCH (t:Topic)<-[:ADDRESSES_TOPIC]-(v:Verse)
WITH t, count(v) AS verse_count

// Find relationships between topics based on co-occurrence in verses
MATCH (t)-[:ADDRESSES_TOPIC]-(v:Verse)-[:ADDRESSES_TOPIC]->(t2:Topic)
WHERE t <> t2
WITH t, t2, count(v) AS shared_verses, verse_count

// Calculate relationship strength
WITH t.name AS topic1, t2.name AS topic2,
     shared_verses, verse_count,
     shared_verses * 1.0 / verse_count AS relationship_strength
WHERE relationship_strength > $strength_threshold

// Calculate network metrics
WITH topic1, collect({topic: topic2, strength: relationship_strength}) AS connections,
     count(*) AS degree
ORDER BY degree DESC

// Return network analysis results
RETURN topic1 AS topic,
       degree AS connection_count,
       connections[0..5] AS strongest_connections
LIMIT $limit

​

Parameters

• strength_threshold: Minimum relationship strength to include (e.g., 0.2)
• limit: Maximum number of topics to return (e.g., 20)

​

Expected Results

​

Variations

• Calculate centrality metrics: WITH topic1, degree, connections, size(connections) * 1.0 / $total_topics AS centrality RETURN topic1, degree, centrality, connections
• Identify thematic communities: This would require implementing community detection algorithms in application code using the relationship data
• Analyze second-degree connections: WITH topic1, connections UNWIND connections AS conn WITH topic1, conn.topic AS intermediate MATCH (t:Topic {name: intermediate})<-[:ADDRESSES_TOPIC]-(v:Verse)-[:ADDRESSES_TOPIC]->(t2:Topic) WHERE t2.name <> topic1 AND t2.name <> intermediate WITH topic1, intermediate, t2.name AS topic2, count(v) AS indirect_strength RETURN topic1, intermediate, topic2, indirect_strength ORDER BY indirect_strength DESC LIMIT 20

​

6. Comparative Analysis of Divine Attributes

​

Description

​

Query

// Find verses addressing divine attributes
MATCH (t:Topic {name: "Divine Attributes"})<-[:ADDRESSES_TOPIC]-(v:Verse)<-[:CONTAINS]-(c:Chapter)

// Extract specific attributes using word roots
MATCH (v)-[:CONTAINS_WORD]->(w:Word)-[:HAS_ROOT]->(r:Root)
WHERE r.text_clean IN $attribute_roots

// Group by revelation period and attribute
WITH c.revelation_place AS place,
     r.text_clean AS attribute_root,
     r.english_trilateral AS attribute_name,
     count(DISTINCT v) AS verse_count,
     collect(DISTINCT v.verse_key)[0..3] AS sample_verses

// Return comparative analysis
RETURN place, attribute_root, attribute_name, verse_count, sample_verses
ORDER BY place, verse_count DESC

​

Parameters

• attribute_roots: List of root words representing divine attributes (e.g., [“رحم”, “علم”, “قدر”, “حكم”])

​

Expected Results

​

Variations

• Calculate relative frequency: WITH place, attribute_root, attribute_name, verse_count, sample_verses, sum(verse_count) OVER (PARTITION BY place) AS total_attributes RETURN place, attribute_root, attribute_name, verse_count, verse_count * 100.0 / total_attributes AS percentage, sample_verses
• Analyze contextual associations: MATCH (v)-[:ADDRESSES_TOPIC]->(other:Topic) WHERE other.name <> "Divine Attributes" WITH attribute_root, other.name AS associated_topic, count(v) AS co_occurrences RETURN attribute_root, collect({topic: associated_topic, count: co_occurrences})[0..5] AS context
• Compare with explicit names: MATCH (v)-[:CONTAINS_WORD]->(w:Word) WHERE w.text_uthmani IN $divine_names WITH place, w.text_uthmani AS divine_name, count(DISTINCT v) AS name_occurrences RETURN place, divine_name, name_occurrences ORDER BY place, name_occurrences DESC

​

7. Narrative Structure Analysis

​

Description

​

Query

// Find verses related to prophetic narratives
MATCH (t:Topic {name: "Prophets"})<-[:ADDRESSES_TOPIC]-(v:Verse)

// Identify specific prophets mentioned
MATCH (v)-[:ADDRESSES_TOPIC]->(prophet:Topic)
WHERE prophet.name IN $prophet_names

// Identify narrative elements
MATCH (v)-[:ADDRESSES_TOPIC]->(element:Topic)
WHERE element.name IN ["Birth", "Childhood", "Mission", "Challenges", "Resolution", "Lessons"]

// Group by prophet and analyze narrative structure
WITH prophet.name AS prophet_name,
     collect(DISTINCT {verse: v.verse_key, element: element.name, position: v.verse_number}) AS narrative_elements
ORDER BY prophet_name

// Organize narrative elements by position
WITH prophet_name,
     apoc.coll.sortMulti(narrative_elements, ["position"]) AS sorted_elements

// Extract narrative structure
RETURN prophet_name,
       [elem IN sorted_elements | elem.element] AS narrative_structure,
       [elem IN sorted_elements | elem.verse] AS narrative_verses

​

Parameters

• prophet_names: List of prophets to analyze (e.g., [“Noah”, “Abraham”, “Moses”, “Jesus”])

​

Expected Results

​

Variations

• Compare narrative lengths: WITH prophet_name, size(narrative_elements) AS narrative_length RETURN prophet_name, narrative_length ORDER BY narrative_length DESC
• Analyze narrative element distribution: UNWIND narrative_elements AS element WITH element.element AS narrative_element, count(*) AS frequency RETURN narrative_element, frequency ORDER BY frequency DESC
• Identify common narrative patterns: WITH collect(narrative_structure) AS all_structures UNWIND all_structures AS structure WITH structure, count(*) AS frequency WHERE frequency > 1 RETURN structure, frequency ORDER BY frequency DESC

​

8. Contextual Interpretation Analysis

​

Description

​

Query

// Find verses containing a specific concept
MATCH (v:Verse)-[:CONTAINS_WORD]->(w:Word)-[:HAS_ROOT]->(r:Root {text_clean: $concept_root})

// Get thematic context
MATCH (v)-[:ADDRESSES_TOPIC]->(t:Topic)
WHERE NOT t.name IN ["Prophets", "Stories", "Parables"] // Exclude general categories

// Get textual context (surrounding words)
MATCH (v)-[:CONTAINS_WORD]->(context_word:Word)
WHERE abs(context_word.position - w.position) <= 3 AND context_word <> w

// Group by thematic context
WITH t.name AS theme,
     collect(DISTINCT v.verse_key) AS verses,
     collect(DISTINCT context_word.text_uthmani) AS contextual_words

// Calculate contextual statistics
WITH theme, verses, contextual_words,
     size(verses) AS verse_count,
     size(contextual_words) AS vocabulary_size

// Return contextual analysis
RETURN theme,
       verse_count,
       vocabulary_size,
       verses[0..3] AS sample_verses,
       contextual_words[0..10] AS common_context
ORDER BY verse_count DESC
LIMIT $limit

​

Parameters

• concept_root: Root word representing the concept to analyze (e.g., “عدل” for justice)
• limit: Maximum number of thematic contexts to return (e.g., 15)

​

Expected Results

​

Variations

• Analyze semantic variation: MATCH (v:Verse)-[:CONTAINS_WORD]->(w:Word)-[:HAS_ROOT]->(r:Root {text_clean: $concept_root}) WITH v, w, avg(v.embedding) OVER (PARTITION BY t.name) AS theme_centroid, v.embedding AS verse_embedding RETURN t.name AS theme, vector_similarity(theme_centroid, verse_embedding) AS semantic_coherence
• Compare with translation variations: MATCH (v)-[:HAS_TRANSLATION]->(trans:Translation {language_name: "English"}) WITH theme, collect(DISTINCT trans.text) AS translations RETURN theme, [word IN apoc.text.keywords(apoc.text.join(translations, " "), {minSize: 4}) WHERE NOT word IN $stop_words | word] AS translation_keywords
• Identify contextual antonyms: MATCH (v)-[:CONTAINS_WORD]->(opposite:Word)-[:HAS_ROOT]->(opp_root:Root) WHERE opp_root.text_clean IN $opposite_roots WITH theme, opp_root.text_clean AS opposite_concept, count(DISTINCT v) AS co_occurrences RETURN theme, collect({concept: opposite_concept, count: co_occurrences}) AS opposing_concepts