A Practitioner's Guide to Graphs
summary
Speed-run shows a schema-first extractor that turns recipes into triples then into typed nodes with standardized units using an embedding model for fuzzy entity matching, with runnable code and Cypher queries alongside SQL comparisons (Cypher vs SQL). Algorithms demonstrated include personalized PageRank for recommendation-style relevance, shortest-path and K-shortest variants for debugging code, and subgraph matching to find design patterns; a .NET evaluation reports a 40% reduction in tool calls when using graph-context retrieval.
AI-written summary · about
chapters (10)
- 0:00Graphs in AI applications
- 1:44Graph basics speedrun
- 3:05Extracting graphs from text
- 3:46Building graphs with schema
- 5:30Node embeddings standardization
- 6:53Querying ingredient relationships
- 7:55Personalized PageRank importance
- 9:01Shortest path for debugging
- 11:26Subgraph matching patterns
- 13:13Summary and next steps
related talks
55mA Practical Guide to Graph Databases
50mThinking in Relationships: Practical Graph Database Modelling in Java with Neo4j
19mUnraveling the Web of Relationships: Understanding Graph Data
01h 00mGraph Mastery to AI
01h 01mA Skeptics Guide to Graph Databases- 19mUnraveling the Web of Relationships: Understanding Graph Data
55mLet Me Graph That For You
01h 00mWho’s Afraid of Graphs?