Skip to content

Knowledge Graphs

Understanding knowledge graphs is key to using NervaPack effectively.

What is a Knowledge Graph?

A knowledge graph is a network of interconnected entities (nodes) and their relationships (edges). Unlike traditional databases that store data in tables, knowledge graphs represent information as a web of connections.

Example: 

[File: auth.py] ──DEFINES──> [Class: AuthMiddleware]
                                    ↑
                                    │
                            EXPLAINS
                                    │
                            [Doc: Authentication Guide]

Why Knowledge Graphs for Code?

Traditional approaches to code understanding:

❌ Text search — Finds keywords, not concepts
❌ Full-file retrieval — Includes irrelevant code
❌ Chunking — Breaks semantic boundaries

✅ Knowledge graphs — Preserve structural relationships

NervaPack's Graph Model

Node Types

  • file — Source files
  • class — Class definitions
  • function — Function/method definitions
  • import — Import statements
  • markdown — Documentation chunks

Edge Types

  • DEFINES — File defines a class/function/import
  • EXPLAINS — Documentation explains code entity

Graph vs Vector RAG

Approach Retrieval Method Context Quality
Vector RAG Cosine similarity on chunks Often includes irrelevant text
Graph RAG Structural traversal (BFS) Precise, semantically connected

Learn More