🔍 NeoQueryAI : Neo4j and FAISS-Powered AI-driven GraphQL query builder

📌 Project Overview

This project automates GraphQL schema processing and natural language query-to-GraphQL generation by:

• Fetching GraphQL schema via introspection.
• Storing the schema in a Neo4j graph database.
• Extracting & indexing types, fields, queries, and mutations.
• Using LLM-based expansion to add synonyms & enrich search capabilities.
• Building FAISS indexes for efficient similarity search.
• Processing user queries in natural language using LLM + FAISS + Neo4j traversal.
• Generating optimized GraphQL queries dynamically.

---

🚀 Features

✅ GraphQL Schema Extraction – Pulls schema via introspection & stores in Neo4j
✅ FAISS Indexing – Creates FAISS indexes for types, fields, queries, and mutations
✅ LLM-Based Enrichment – Uses GPT-4 Turbo to expand schema with synonyms & context
✅ Smart FAISS Search – Matches user queries to schema fields with high accuracy
✅ Query Path Selection – Finds optimal queries & traversal paths in Neo4j
✅ GraphQL Query Generation – Converts results into executable GraphQL queries
✅ Minimal Query Optimization – Ensures efficient & minimal query selection

---

🛠 Installation

1️⃣ Clone the Repository

git clone https://github.com/saurabhbansal123/NeoQueryAI.git
cd NeoQueryAI

2️⃣ Install Dependencies

Ensure Python 3.8+ is installed, then run:

pip install -r requirements.txt

3️⃣ Configure Environment

Use the example configuration:

Edit config/config.ini and set:

• GraphQL API Endpoint & Access Token
• Neo4j Credentials
• OpenAI API Key

---

⚙️ Configuration

Update config/config.ini:

[graphql]
graphql_api_url = https://your-graphql-endpoint.com/graphql
graphql_access_token = your_graphql_api_key

[neo4j]
uri = bolt://localhost:7687
username = neo4j
password = your_neo4j_password

[openai]
openai_api_key = your_openai_api_key

---

📌 Workflow & Execution

1️⃣ Fetch & Store GraphQL Schema

Run:

python Process_data.py

This will:

• Fetch schema via introspection
• Store schema as a graph in Neo4j
• Process schema into a structured vocabulary
• Build FAISS indexes

2️⃣ Process User Queries & Generate GraphQL

Run:

python GenerateGraphQL.py

This will:

• Use LLM to extract structured fields
• Search for matching GraphQL fields in FAISS
• Identify queries from Neo4j schema
• Optimize query selection & build GraphQL

---

🔍 How It Works

1️⃣ Schema Processing & Enrichment

1. Extracts types, fields, queries, mutations from GraphQL API.
2. Stores schema as a graph in Neo4j.
3. Expands vocabulary with GPT-4 Turbo (synonyms & context).
4. Indexes schema using FAISS for fast similarity search.

2️⃣ FAISS-Powered Query Processing

1. User inputs a natural language query (e.g., "Fetch event details").
2. LLM extracts relevant GraphQL fields, types, and queries.
3. FAISS searches for best schema matches.
4. Neo4j finds optimal queries & traversal paths.

3️⃣ Optimized GraphQL Generation

1. Prioritizes minimal depth & efficient query selection.
2. Builds GraphQL query dynamically.
3. Returns ready-to-use GraphQL query.

---

🔥 Example: Generating a GraphQL Query

User Input

"Fetch event details along with attendees and their emails"

Extracted Entities & Fields

{
  "queries": ["Query.eventsConnection", "Query.attendeesConnection"],
  "fields": ["Event.id", "Event.name", "Attendee.firstName", "Attendee.email"]
}

Generated GraphQL Query

query {
  eventsConnection {
    nodes {
      id
      name
      attendees {
        nodes {
          firstName
          email
        }
      }
    }
  }
}

---

📜 License

This project is licensed under the Apache 2.0 License.

You may:

• Use, modify, and distribute the software for any purpose.
• Create derivative works and incorporate them into other projects.
• Commercially use the software.

You must:

• Provide proper attribution by crediting the original authors when using this work in derivative or redistributed projects.
• Include a copy of this license in any distribution.
• State any modifications you make to the original source code.
• Not hold the original authors liable for any issues arising from the use of this software.

For full details, see: Apache 2.0 License.

---

🤝 Contributing

💡 Found a bug? Want to improve this? Contributions are welcome!

Some opportunity areas are already highlighted in the code with a TODO comment.

Steps to Contribute

1. Fork the repository
2. Create a new branch

   git checkout -b feature-branch

3. Commit your changes

   git commit -m "Added new feature"

4. Push your branch

   git push origin feature-branch

5. Create a Pull Request (PR)

---

📧 Contact

For questions or collaboration:

• GitHub Issues: Open an Issue
• Email: [email protected]