semantica/cookbook/introduction/07_Building_Knowledge_Graphs.ipynb

Building Knowledge Graphs

Overview

This notebook demonstrates how to build knowledge graphs from entities and relationships using Semantica's graph building modules. You'll learn to use GraphBuilder, EntityResolver, and GraphValidator.

Documentation: API Reference

Learning Objectives

• Use GraphBuilder to construct knowledge graphs
• Use EntityResolver to resolve entity conflicts
• Use GraphValidator to validate graph structure Note: For deduplication, use the semantica.deduplication module.

Installation

Install Semantica from PyPI:

pip install semantica
# Or with all optional dependencies:
pip install semantica[all]

---

Step 1: Build Knowledge Graph

Construct a knowledge graph from entities and relationships.

In [ ]:

from semantica.kg import GraphBuilder
from semantica.semantic_extract import NERExtractor, RelationExtractor

builder = GraphBuilder()
ner_extractor = NERExtractor()
relation_extractor = RelationExtractor()

text = "Apple Inc. is a technology company. Tim Cook is the CEO of Apple Inc. Apple Inc. is headquartered in Cupertino, California."

entities_list = ner_extractor.extract(text)
relationships_list = relation_extractor.extract(text, entities_list)

entities = []
for i, entity in enumerate(entities_list[:5], 1):
    entities.append({
        "id": f"e{i}",
        "type": entity.get("type", "Entity"),
        "name": entity.get("text", entity.get("entity", "")),
        "properties": {}
    })

relationships = []
for i, rel in enumerate(relationships_list[:3], 1):
    relationships.append({
        "source": f"e{1}",
        "target": f"e{i+1}",
        "type": rel.get("type", "related_to"),
        "properties": {}
    })

knowledge_graph = builder.build(entities, relationships)

print(f"Built knowledge graph with {len(knowledge_graph.get('entities', []))} entities")
print(f"Relationships: {len(knowledge_graph.get('relationships', []))}")

Step 2: Entity Resolution

Resolve entity conflicts and duplicates.

In [ ]:

from semantica.kg import EntityResolver

entity_resolver = EntityResolver()

resolved_entities = entity_resolver.resolve(entities)

print(f"Original entities: {len(entities)}")
print(f"Resolved entities: {len(resolved_entities)}")

Step 3: Graph Validation

Validate the knowledge graph structure.

In [ ]:

from semantica.kg import GraphValidator

graph_validator = GraphValidator()

validation_result = graph_validator.validate(knowledge_graph)

print(f"Graph validation: {validation_result.get('valid', False)}")
print(f"Issues: {len(validation_result.get('issues', []))}")

Step 4: Deduplication

Remove duplicate entities from the graph.

In [ ]:

from semantica.deduplication import DuplicateDetector, EntityMerger, MergeStrategy

# Detect duplicates
detector = DuplicateDetector(similarity_threshold=0.8)
duplicate_groups = detector.detect_duplicate_groups(knowledge_graph.get('entities', []))

# Merge duplicates
merger = EntityMerger()
merge_operations = merger.merge_duplicates(
    knowledge_graph.get('entities', []),
    strategy=MergeStrategy.KEEP_MOST_COMPLETE
)

deduplicated_entities = [op.merged_entity for op in merge_operations]

print(f"Original entities: {len(knowledge_graph.get('entities', []))}")
print(f"Deduplicated entities: {len(deduplicated_entities)}")

Summary

You've learned how to build knowledge graphs:

• GraphBuilder: Construct knowledge graphs from entities and relationships
• EntityResolver: Resolve entity conflicts and duplicates
• GraphValidator: Validate graph structure and quality
• Deduplication: Use semantica.deduplication module for removing duplicate entities

Next: Learn how to analyze graphs in the Graph_Analytics notebook.