concept

Neo4j LLM Knowledge Graph Builder

Facts (10)

Sources

How to Improve Multi-Hop Reasoning With Knowledge Graphs and ... neo4j.com Neo4j Jun 18, 2025 10 facts

claimThe Neo4j LLM Knowledge Graph Builder is an online application that transforms unstructured content, such as PDFs, documents, URLs, and YouTube transcripts, into structured knowledge graphs stored in Neo4j.

procedureThe Neo4j LLM Knowledge Graph Builder processes text into a graph in eight steps: (1) Uploaded sources are stored as Document nodes in the graph. (2) Document types are loaded with LangChain loaders. (3) The content is split into Chunks. (4) Chunks are stored in the graph and connect to the document and to each other for advanced RAG patterns. (5) Highly similar chunks are connected with a SIMILAR relationship (with a weight attribute) to form a k-nearest neighbors graph. (6) Embeddings are computed and stored in the chunks and vector index. (7) The llm-graph-transformer or diffbot-graph-transformer extracts entities and relationships from the text. (8) Entities and their relationships are stored in the graph and connect to the originating chunks.

referenceKey features of the Neo4j LLM Knowledge Graph Builder include multimodal ingestion (PDFs, HTML, transcripts, URLs, cloud buckets), schema configuration (preset, existing Neo4j schema, or LLM-inferred), integrated vector and graph retrieval for hybrid search, visualization via in-app viewer or Neo4j Bloom, and flexible deployment options (Neo4j Aura or local Docker Compose).

claimThe Neo4j LLM Knowledge Graph Builder allows users to convert unstructured content, such as PDFs, transcripts, or webpages, into structured graphs for use in retrieval-augmented applications.

claimThe Neo4j LLM Knowledge Graph Builder integrates large language models, including OpenAI, Gemini, Claude, and Diffbot, with Neo4j’s graph-native storage and retrieval capabilities.

claimThe Neo4j LLM Knowledge Graph Builder utilizes a React frontend and a Python FastAPI backend, which can be deployed on Google Cloud Run or locally using Docker Compose.

claimThe Neo4j LLM Knowledge Graph Builder creates a lexical graph of documents and chunks with embeddings, alongside an entity graph containing nodes and relationships, both stored in a Neo4j database.

claimUsers can employ retrieval-augmented generation (RAG) approaches, specifically GraphRAG, vector search, and Text2Cypher, to query data within the Neo4j LLM Knowledge Graph Builder.

claimThe Neo4j LLM Knowledge Graph Builder uses the llm-graph-transformer module, which Neo4j contributed to LangChain, and supports other LangChain integrations for tasks like GraphRAG search.

claimThe Neo4j LLM Knowledge Graph Builder links every graph element to its source text and chunk, providing explainability for the generated outputs.