MazzaWill/neo4j-python-pandas-py2neo-v3

Excel-to-Neo4j knowledge graph examples: legacy py2neo v3 plus modern Neo4j GraphRAG/vector search.

What it solves

This project provides examples and tools for converting tabular data from Excel spreadsheets into Neo4j knowledge graphs. It addresses the challenge of transforming structured spreadsheet data into a graph database for better relationship mapping, semantic retrieval, and downstream machine learning experiments.

How it works

The project uses pandas to read and extract node and relationship data from Excel files. It then populates a Neo4j database using either a legacy py2neo v3 path for educational purposes or a modern Neo4j driver path. The modern implementation includes vector indexes and GraphRAG-style semantic retrieval to enable advanced querying of the graph data.

Who it’s for

  • Data Engineers looking to migrate tabular data into graph databases.
  • Developers implementing GraphRAG or vector search within Neo4j.
  • Learners studying the process of knowledge graph construction from structured data.

Highlights

  • Dual-Path Implementation: Supports both legacy py2neo v3 for educational baselines and modern Neo4j 5+ drivers for production-ready GraphRAG.
  • GraphRAG Integration: Includes examples of semantic retrieval using vector indexes and embeddings.
  • ML Readiness: Features a utility to convert Neo4j relationship data into matrices for machine learning models.
  • AI Agent Skill: Provides a specialized skill for AI coding agents to help them design knowledge graphs from CSV/Excel data and generate Cypher queries.

Related

  • Project
    Helldez/BigMoeOnEdge

    An inference engine that enables running Mixture-of-Experts models larger than a device's RAM by streaming experts from flash storage in real-time.

  • Project
    Fausto-404/js-reverse-automation--skill

    An automated frontend JS reverse engineering framework that uses AI and JSRPC to turn real browsers into decryption engines and automatically deploys Flask and Burp Suite proxies.

  • Project
    flashrt-project/FlashRT

    A high-performance real-time inference engine for small-batch AI workloads that uses hand-written kernels and static graphs to minimize latency without requiring model compilation.

  • Project
    nadimtuhin/claude-token-optimizer

    A CLI tool that optimizes Claude Code's context window by restructuring project documentation into a lean, demand-driven loading system, reducing startup token waste.