Quick Start¶

Build your first knowledge graph in 5 minutes using the REPL (interactive shell).

Step 1: Launch the REPL¶

network-rag repl

Step 2: Connect to a Database¶

[no db]> db connect papers.db
✓ Connected to papers.db

Step 3: Add Documents¶

papers.db> add "Attention Is All You Need - introducing Transformer architecture"
✓ Added document: abc123...

papers.db [1 docs]> add "BERT: Bidirectional Encoder Representations from Transformers"
✓ Added document: def456...

papers.db [2 docs]> add "GPT-3: Language Models are Few-Shot Learners"
✓ Added document: ghi789...

Step 4: Build the Network¶

papers.db [3 docs]> build
ℹ Fitting TF-IDF vectorizer...
ℹ Building network...
✓ Network built: 3 nodes, 2 edges

Step 5: Explore the Network¶

papers.db [3 docs, 2 edges]> graph info

Network Statistics
========================================
Nodes: 3
Edges: 2
Density: 0.667
Average degree: 1.3

Step 6: Search¶

papers.db [3 docs, 2 edges]> search "transformers for NLP"
Search: transformers for NLP
Found 3 results:

1. abc123 (score: 0.892)
   Attention Is All You Need - introducing Transformer architecture

2. def456 (score: 0.745)
   BERT: Bidirectional Encoder Representations from Transformers

3. ghi789 (score: 0.621)
   GPT-3: Language Models are Few-Shot Learners

You Did It! 🎉¶

You just: - ✅ Created a knowledge graph - ✅ Built a similarity network - ✅ Performed topology-aware retrieval

Next Steps¶

Learn the Concepts¶

Core Concepts - Understand how it works
Network Topology - Why graphs matter

Use Structured Documents¶

Instead of plain text, use structured JSON documents:

papers.db> add {"title": "Attention Is All You Need", "abstract": "...", "tags": ["NLP", "transformers"]}

See: YAML DSL Guide for field-specific similarity

Try Other Interfaces¶

Python API - Programmatic access
CLI Reference - Command-line usage

Complete Tutorials¶

Research Papers - Build a paper knowledge graph
E-commerce - Product recommendations
RAG Pipeline - Full LLM integration