Knowledge graphs --enter- the Hype Cycle
summary
Claims a working fashion knowledge graph for Farfetch that links products to taxonomy nodes and powers search, recommendations, ranking and personalization. Uses Python tooling (NetworkX demos), Neo4j plus cloud options like AWS Neptune, and ML: computer vision, deep learning, NLP, graph embeddings and human-in-the-loop curation. Dataset size exceeds over a million nodes with many more relationships. Demo code and site generator live on GitHub and the approach reports practical improvements in discovery rather than vague hype.
AI-written summary · about
chapters (14)
- 0:00Welcome
- 1:09Hype Cycle explained
- 3:43Why knowledge graphs matter
- 8:03What is a knowledge graph
- 13:34Graphs in daily life
- 15:01Panama Papers with graphs
- 19:54Fashion knowledge graph
- 22:02Fashion taxonomy
- 27:01Graph construction techniques
- 30:48Graph queries demo
- 33:02Stakeholder communication
- 34:51Why use knowledge graphs
- 36:16Graphs in AI/ML
- 38:03Q&A
related talks
38mKnowledge Graphs : Adding a New Dimension to Vector Search and Generative AI
30mHow Graph Data Science Can Turbocharge Your Knowledge Graph
26mGraph Databases will change your (freakin') life
27mAdaptive Knowledge Graphs for Dynamic RAG Systems
22mThe Nexus of Innovation: How Knowledge Graphs Fuel AI Scientific Breakthroughs
30mGraph Data Science
28mRole of Knowledge Graphs in Mental Health Diagnosis and Cure
49mModern Graph Databases