Practical overview
When it fits
Use it to explore a document-to-knowledge-graph workflow or demonstrate natural-language querying over a graph built from SEC filings.
Who it is for
- Developers exploring Bedrock-based document extraction into Neo4j.
- Data teams prototyping knowledge graphs from regulatory filings.
- People demonstrating graph queries in natural language to non-technical users.
Documented uses
- Extract entities and relationships from SEC EDGAR documents.
- Ingest extracted information into a Neo4j database.
- Ask natural-language questions that Bedrock converts into Cypher queries against the graph.
Requirements & limits
- The notebook setup is described for SageMaker Studio, while the Streamlit interface is described for an EC2 virtual machine.
- The workflow depends on a Neo4j instance and AWS IAM permissions for Amazon Bedrock.
- The repository lists an Apache-2.0 license.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Prepare the SageMaker notebook workflow
Set up the documented notebook environment for the SEC EDGAR ingestion and knowledge-graph demonstration.
Before you start
- A SageMaker Studio environment.
- A Neo4j instance for the graph data.
- An AWS identity with IAM permissions to call Amazon Bedrock.
Steps
Clone the repository into SageMaker Studio.
Run notebooks numbered 1 and 2.
Use the notebook environment's AWS identity to access Bedrock and connect the workflow to a Neo4j instance.
Expected outcome
A prepared notebook workflow for exploring extraction of entities and relationships from EDGAR files and their ingestion into Neo4j.
Guide limitations
- The guide does not provide a complete notebook execution walkthrough.
- A Neo4j instance and Bedrock permissions must be arranged separately.
Set up the Streamlit chat interface on EC2
Prepare the documented Streamlit application environment on an EC2 virtual machine.
Before you start
- An EC2 virtual machine with HTTP access on the port used by the application.
- AWS CLI credentials and permission to connect to the instance.
- Bedrock access and a Neo4j instance, with the corresponding values available for the application secrets.
Steps
Connect to the EC2 instance using AWS CLI and install the documented Git, Python, pip, and virtual-environment tools.
Clone the repository, enter the ui directory, and install dependencies from requirements.txt.
Copy secrets.toml.example to secrets.toml and configure the Bedrock and Neo4j settings and credentials.
Start the interface with the documented Streamlit command on the chosen port.
Expected outcome
A locally deployed Streamlit dashboard and chat interface configured to connect to Bedrock and Neo4j.
Guide limitations
- The documented example uses port 80, which requires root access and an instance network rule allowing HTTP traffic.
- The instructions describe deployment setup; they do not establish compatibility with other deployment environments.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 18 October 2026
- Original website title
- neo4j-partners/neo4j-generative-ai-aws · Official homepage title · retained in its source language