neo4j-labs/llm-graph-builder
Llm Graph Builder
llm-graph-builder is a tool for transforming unstructured data into structured Neo4j graph databases using large language models. It functions as a graph orchestrator that automates the construction of nodes and relationships from raw text based on custom schemas.
The project provides a visualizer for analyzing relational data as interactive networks and a token monitor to track daily and monthly API consumption per user. It also includes a vector embedding generator that utilizes configurable model providers to enable semantic search and retrieval augmented generation.
The system covers capabilities for unstructured data analysis, conversational data retrieval through natural language interfaces, and semantic indexing via vector embeddings.
Features
- Knowledge Graph Construction Tools - Transforms unstructured data into structured knowledge graphs using large language models and custom schemas.
- Vector Embeddings - Generates vector representations of data using various configurable embedding model providers.
- Graph Database Integrations - Integrates large language models with Neo4j to automate the population of graph data structures.
- Graph Population Automators - Automates the construction of nodes and relationships within Neo4j from raw text.
- Natural Language Querying - Implements the translation of natural language user inputs into structured graph queries for record retrieval.
- Knowledge Graph Builders - Builds structured Neo4j graph databases from unstructured data using LLMs and custom schemas.
- Unstructured Data Transformation Tools - Converts raw narrative text into queryable graph formats to uncover hidden entity connections.
- Vector Indexing - Converts textual data into numerical embeddings to enable semantic similarity searches within the graph.
- Graph Schema Extractions - Uses language models to transform unstructured text into structured nodes and relationships based on defined schemas.
- Multi-Provider Abstractions - Abstracts multiple embedding model providers into a common interface for generating vector representations.
- Natural Language Query Interfaces - Provides a natural language chat interface for users to query database records and retrieve source metadata.
- Graph Visualizers - Provides a visual interface for exploring complex relationships between data entities in a knowledge graph.
- Conversational Retrieval - Enables an interactive dialogue grounded in retrieved graph database records and source metadata.
- Vector-Store Augmented Generation - Implements vector-store augmented generation to enable semantic retrieval within the graph.
- Interactive Graph Visualizers - Renders extracted nodes and edges as interactive visual networks for relationship analysis in the browser.
