Rig-Neo4j Integration

The rig-neo4j crate provides a vector store implementation using Neo4j as the underlying datastore. This integration allows for efficient vector-based searches within a Neo4j graph database, leveraging Neo4j’s vector index capabilities.

Key Features

• Vector Indexing: Supports creating and querying vector indexes in Neo4j, enabling semantic search capabilities.
• Integration with OpenAI: Utilizes OpenAI’s embedding models to generate vector representations of data.
• Flexible Querying: Offers methods to perform top-N searches and retrieve node IDs based on vector similarity.
• Customizable Index Configuration: Allows configuration of index properties such as dimensions and similarity functions.

Prerequisites

• Neo4j GenAI Plugin: Required for vector indexing capabilities. Enabled by default in Neo4j Aura or can be installed on self-managed instances.
• Pre-existing Vector Index: The Neo4j vector index must be created before performing searches. This can be done using the Neo4j browser, Cypher queries, or the Neo4jClient::create_vector_index method.

Usage

Setup

Add the rig-neo4j crate to your Cargo.toml:

[dependencies]
rig-neo4j = "0.2.0"

Example Workflow

1. Connect to Neo4j: Establish a connection using the Neo4jClient.

let neo4j_client = Neo4jClient::connect("neo4j://localhost:7687", "username", "password").await?;

2. Create Vector Index: Define and create a vector index on your data.

neo4j_client.create_vector_index(
    IndexConfig::new("moviePlots"),
    "Movie",
    &model
).await?;

3. Perform Vector Search: Query the vector index for similar nodes.

let results = index.top_n::<Movie>("a historical movie on quebec", 5).await?;

Example Code

#[tokio::main]
async fn main() -> Result<(), anyhow::Error> {
    let neo4j_client = Neo4jClient::connect("neo4j://localhost:7687", "username", "password").await?;
    let model = openai_client.embedding_model(TEXT_EMBEDDING_ADA_002);
 
    neo4j_client.create_vector_index(
        IndexConfig::new("moviePlots"),
        "Movie",
        &model
    ).await?;
 
    let index = neo4j_client.get_index(model, "moviePlots", SearchParams::default()).await?;
    let results = index.top_n::<Movie>("a historical movie on quebec", 5).await?;
    println!("{:#?}", results);
 
    Ok(())
}

Additional Resources

• Examples: See the examples directory for detailed usage scenarios.
• Neo4j Documentation: Refer to the Neo4j vector index documentation for more information on setting up and using vector indexes.