Intelligence · May 2025 — Jun 2025

Entity Knowledge Graph

Build a Living Knowledge Graph from Your Documents

Entity Knowledge Graph

Automatically extract entities and relationships from documents to build a living knowledge graph.

Services

  • Entity Extraction
  • Graph Building
  • Dossier Generation

Technologies

  • Neo4j
  • SurrealDB
  • LLM Gateway

Software

  • Python
  • FastAPI
  • LiteLLM

Client

CogMem Open Source

Introduction

CogMem automatically extracts named entities — people, projects, concepts, tools, places — from ingested documents and builds a knowledge graph of relationships between them. Entity Dossier workflows generate Wikipedia-style pages with claims and evidence for any entity in the graph.

Challenges

  • Extracting entities accurately from diverse document types.
  • Resolving entity aliases and deduplication across sources.
  • Building meaningful relationships between entities.
  • Generating useful dossiers from claims and evidence.

Solution

The EntityExtractAgent uses LLM-powered extraction to identify entities from passages. Entities are linked into a knowledge graph with typed relationships. The Entity Dossier workflow gathers all claims about an entity, hydrates supporting evidence, and renders a comprehensive dossier artifact.

Work Done

Entity Extraction

LLM-powered extraction of people, projects, concepts, and places.

Knowledge Graph

Typed relationships between entities with graph traversal.

Entity Dossiers

Wikipedia-style pages generated from claims and evidence.

Alias Resolution

Canonical entity representations with alias tracking.

Cross-Source Linking

Entities are linked across documents and data sources.

Inspectable Graph

Full visibility into entity relationships and evidence trails.

Ready to Build Smarter Knowledge Systems?

Get started with CogMem and transform how your AI agents remember, organize, and evolve knowledge.

Get Started

Contact Us

Get in Touch

Have questions about CogMem? We'd love to hear from you.

Website

cogmem.ai