Graph Algorithms: Predict Real-World Behavior
summary
Shows that averages fail for real networks and argues real graphs follow power law distributions, not bell curves. Uses Neo4j graph algorithms (PageRank, betweenness, closeness, modularity, triangle count) and demos analysis on a Game of Thrones dataset, computing betweenness and PageRank, storing scores as node properties and comparing book-to-book influence. Presents concrete use cases — call routing, fraud rings, electrical grid where removing four nodes slows cascades — and features practical tools like cipher projection and Neo4j Sandbox for reproducible experiments.
AI-written summary · about
chapters (10)
- 0:00Introduction to speaker
- 1:01The problem with averages
- 2:14Graph theory foundations
- 5:55Pathfinding algorithms
- 9:53Centrality algorithms
- 16:07Community detection
- 20:08Motivation for Neo4j
- 21:45Neo4j ecosystem and tools
- 24:55Game of Thrones demo
- 40:56Learning resources
related talks
43mGraphs are Everywhere
46mGraph Databases: Talking about your Data Relationships with Python
29mThinking in Graphs – An Intro to Graph Databases & Applications
32mUnderstanding Complexly Connected Data using Graph Databases
01h 38mNetwork Science for Fun and Profit
47mWhat Business Can Learn from Dating
55mLet Me Graph That For You
31mGraph Analytics - From the Whiteboard to Your Toolbox