Nyah Macklin explains why finding similar text is different from retrieving connected context. Knowledge graphsA knowledge graph represents entities, concepts and facts as connected nodes and relationships that software can query and traverse. represent entities and relationships that an agent can traverse for multi-hop questions and aggregations. The workshop combines that structure with vector searchVector search finds items by comparing numerical representations of their features or meaning. rather than treating either method as a complete solution.
Nyah Macklin and Jeremy Adams - Casañas walk through document chunking, typed entity and relationship extraction, graph constructionGraph retrieval-augmented generation uses connected entities and relationships to supply context for an AI-generated response. and a Neo4j Aura agent. Jeremy Adams - Casañas handles the Python setup and query demonstration, showing how structured Cypher queries and a vector index provide different retrieval pathsRetrieval is the process of selecting relevant stored information and returning it to an AI system for the current task..
A counting example produces an incorrect result, prompting discussion of repeated ingestion and the state of the graph. The demonstration makes the architecture tangible without establishing a blanket accuracy improvement: teams still need to validate extraction, avoid duplicates and check query results. Promotional course, account, event and giveaway invitations are omitted.
Watch on YouTube
