Text-attributed graph

As explained in Knowledge Graphs and LLMs in Action

A text-attributed graph is a graph whose entities or relationships have associated textual attributes.

A text-attributed graph is a graph whose nodes and relationships carry textual attributes. Those attributes can preserve descriptions, extracted content, or other text alongside structured connections. section 13.4.3

Use the structure to support retrieval and grounding:

• Combine textual attributes with graph metadata such as publication date, document type, source, and author. section 13.4.3
• Connect the graph to Graph RAG so typed relationships and condensed knowledge provide context for an LLM. section 13.4.3
• Distinguish it from a text-paired graph, which associates nodes, relationships, and graphs with their source documents. section 13.4.3

Archive knowledge graph

The Rockefeller Archive Center knowledge graph combines text-attributed and text-paired structures: documents become nodes with author, type, and date properties, while extracted entities and relationships remain traceable to their sources. section 13.4.3

Figure 13.3 The Rockefeller Archive Center KG combines a text-attributed graph and a text-paired graph. Documents are represented as nodes, each with properties representing metadata such as author, type of document, and date. The extracted entities and relationships are traceable back to the original documents, which enables us to design a KG-based document selection tool for AI agents.

Trace how document nodes, metadata, extracted entities, and source relationships support graph-based document selection. section 13.4.3

How Knowledge Graphs and LLMs in Action covers it

How deep it goes

Defines textual attributes on nodes and relationships, then applies them to metadata-aware retrieval without detailing graph construction internals.

Examples

Uses the Rockefeller Archive Center knowledge graph, document metadata, extracted entities, and source-traceable relationships.

What you can do after

Apply it via Graph RAG to retrieve grounded subgraphs or documents for LLM-generated responses.

How it connects

Connects knowledge graphs, text-paired graphs, Graph RAG, LLMs, document metadata, and vector search.

liveBookmanning.com

Read Knowledge Graphs and LLMs in Action on liveBook Get the book at manning.com

This explainer was generated by Manning's knowledge graph from the text of Knowledge Graphs and LLMs in Action. The section links point to the parts of the book it draws on.