Graph ML – The Next Level Of Machine Learning; Learn How To Use It And When Not To
summary
Claims that graphs boost fraud detection, ranking (PageRank), recommendations and chemistry by treating relationships as first‑class data and noting about 30% of ML papers now mention graph approaches. Covers practical tradeoffs: ArangoDB demo Wrangleflix uses graph embeddings or collaborative filtering versus full analytics; graph neural networks provide inductive models that train offline and give fast inference while analytics often requires transductive, full‑graph recomputation and scales poorly as graph size and update frequency grow.
AI-written summary · about
chapters (7)
- 0:00Introduction to graph ML
- 2:10Graph databases explained
- 3:52The evolution of graph data
- 6:56Graph machine learning basics
- 10:42Algorithms and use cases
- 13:25Choosing the right approach
- 20:24Conclusion and summary
related talks
01h 06mMachine Learning over Graphs
43mAnalytics and AI Applied to Graph Databases - Introduction and Demo
50mMachine Learning + Graph Databases for Better Recommendations
12mMake ML outcomes roar with graph neural networks on TigerGraph and AWS (PRT097)
31mGraph-Based Machine Learning with Python
26mAdvancing Financial Fraud Detection with Graph Dbs
56mGraph Analytic Systems
26mGraph Databases will change your (freakin') life