Building Knowledge Graphs
summary
Claims that knowledge graphs power data catalogs/fabric and MDM, cut 30% of time lost to finding data (McKinsey), and already link ~31 billion triples in Linked Open Data. It contrasts LPG (Neo4j style) for traversal with RDF/OWL/SPARQL for semantics, provenance (PROV‑O) and temporality, shows a working graph RAG that beats plain RAG by using ontologies, and describes integrations with Elastic, vector DBs, Spark/Kafka and a GraphDB v10.8 agentic engine.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 7:55Graphs in daily life
- 15:00Knowledge graph vs graph
- 22:38Semantic search with graphs
- 28:03RDF vs property graphs
- 33:42Building graphs from data
- 38:43Graphs and LLMs synergy
- 40:05Q&A: Adoption barriers
- 42:12Q&A: Data changes handling
- 44:19Q&A: Modeling in RAG
related talks
44mBuilding Knowledge Graphs- 44mBuilding Knowledge Graphs
24mWhy You Need a Knowledge Graph And How to Build It
30mGraph Data Science
21mMastering Knowledge Graph Transformations Without Compromising Meaning
39mEnhancing RAG with Neo4j Knowledge Graph
15mGraphRAG-Powered AI Agents: Transforming Corporate Metadata into Intelligent Knowledge
38mKnowledge Graphs : Adding a New Dimension to Vector Search and Generative AI