Introduction to Docling Graph

Welcome to Docling Graph! This section introduces the core concepts of knowledge graph extraction from documents.

What is Docling Graph?

Docling Graph transforms unstructured documents into validated knowledge graphs with precise semantic relationships. Unlike traditional approaches that convert text to embeddings (losing exact relationships), Docling Graph preserves explicit entity connections.

Key Advantage: Answer questions like "who issued what to whom" with exact relationships, not approximate embeddings.

Why Knowledge Graphs?

Knowledge graphs are essential for domains requiring exact entity connections:

• Chemistry: Track compounds, reactions, and measurements
• Finance/Legal: Map instruments, obligations, and dependencies
• Research: Connect authors, methodologies, and results
• Healthcare: Relate patients, treatments, and outcomes

How It Works

Docling Graph follows a clear pipeline:

1. Installation - Set up environment
2. Schema Definition - Create Pydantic templates
3. Pipeline Configuration - Configure backend and mode
4. Extraction Process - Run extraction
5. Graph Management - Export and visualize

Quick Example

from docling_graph import run_pipeline, PipelineConfig
from pydantic import BaseModel, Field

class Person(BaseModel):
    model_config = {'is_entity': True, 'graph_id_fields': ['name']}
    name: str = Field(description="Person's full name")
    email: str = Field(description="Email address")

config = PipelineConfig(
    source="document.pdf",
    template=Person,
    backend="llm",
    inference="remote"
)

context = run_pipeline(config)
print(f"Extracted {context.knowledge_graph.number_of_nodes()} nodes")

Learn More

• Key Concepts - Entities, components, nodes, edges
• Use Cases - Domain-specific examples
• Architecture - System design
• Quick Start - 5-minute tutorial

---

Ready to start? Begin with installation or try the quick start.