PearsonOLT_GraphRAGFundamentals_Oct2025.pdf

In this fast-paced technology industry, it is easy to feel overwhelmed by the amount of learning needed to simply keep up. Start with basic concepts of the AI industry, and then dive into technologies that support the incredible speed and developments in this space from the Java perspective. Vectors, RAG, GraphRAG, agents, MCP, and whatever comes next fit together as puzzle pieces … all you have to know is when and how to use them.
These concepts are not as complex as they seem, and once you learn the foundations, everything else (including new developments) builds on top of those blocks. This course will help you learn the terminology, see how it works through live demos, and then get hands-on experience with specific tools for building GenAI apps. We will cover vectors, RAG, and GraphRAG in detail, as well as highlight recent trends and developments in the space. Get caught up on the GenAI industry and open opportunities for deeper exploration, innovation, and just-right solutions!
Code: https://github.com/JMHReif/graphrag-fundamentals

Transcript

1. Click to edit Master title style GraphRAG Fundamentals Improve Your GenAI Solutions with Connected Data Jennifer Reif Neo4j Developer Advocate, Java
2. POLL! How familiar are you with graph databases? Select one. • Very familiar • Somewhat familiar • A little familiar • Not at all familiar • What is a graph database?
3. About Jennifer Developer advocate, Neo4j • Java champion • Tech speaker/blogger/podcaster/author • Website+Blog: jmhreif.com • Github: github.com/JMHReif • Email: jennifer.reif@neo4j.com • LinkedIn: linkedin.com/in/jmhreif • X: x.com/JMHReif
4. Outline and Resources • Notes: bit.ly/jmhreifPearson • Outline: • Graph databases + Java apps • GenAI + RAG + GraphRAG • Building GraphRAG solutions Notes: bit.ly/jmhreifPearson
5. Layers …of systems with GenAI • More layers = better result • Complexity vs value Notes: bit.ly/jmhreifPearson
6. Layers of systems Today’s step-by-step Traditional systems neo4j Vector search / naive RAG GraphRAG Agents / MCP Notes: bit.ly/jmhreifPearson
7. Introduction to Graph Databases
8. Social network LinkedIn Jennifer Priya ?? Notes: bit.ly/jmhreifPearson
9. Social network Explicit links Company Jennifer Priya WORKED_FOR WORKED_FOR
10. Social network Explicit links Company Jennifer Priya School W ORKED_FOR WORKED_FOR ATTENDED ATTENDED
11. Skill Degree Company Jennifer Priya School ATTENDED ATTENDED W ORKED_FOR WORKED_FOR H AS_SKILL COM PLETED Degree COM PLETED HAS_SKILL Social network Explicit links
12. Dominic Skill Company Jennifer Priya School ATTENDED ATTENDED W ORKED_FOR WORKED_FOR H AS_SKILL HAS_SKILL Degree C O M PLETED COM PLETED Degree HAS_SKILL Social network Explicit links COM PLETED
13. Person Company Person Person School ATTENDED ATTENDED W ORKED_FOR WORKED_FOR Degree C O M PLETED COM PLETED Work, Inc. Priya Jennifer Hire Me U CMIS Dominic Degree HAS_SKILL Skill H AS_SKILL HAS_SKILL Java CS Social network Explicit links COM PLETED
14. What is a graph? Answers through relationships • How many coworkers shared skills/degrees? • Who is within 2 connections of me and knows Neo4j? • Who adds new skills? • Who might be a good fi t for this project? Edward Jones Adrian Person Company Person Person School ATTENDED ATTENDED W ORKED_FOR WORKED_FOR Degree C O M PLETED COM PLETED Priya Jennifer CMIS Degree HAS_SKILL Skill H AS_SKILL HAS_SKILL Java CS Dominic Work, Inc. Hire Me U COM PLETED
15. Skill Skill Skill School School Degree Degree Company Company Company Company Person Person Person Person Person Person Person Person Person Company Person Work, Inc. Priya Dominic Degree Degree CMIS Skill Java CS Non-graph social network Implicit links Person Jennifer School Hire Me U
16. Nodes (vertices) Objects or entities • Still have entities • Flexible schema • Retain uniqueness Person Skill Company Person Person School Degree Priya Jennifer CMIS Java Degree CS Dominic Work, Inc. Hire Me U Username: jmhreif
17. Relationships (edges) Connect entities • Equal importance • Paved path > cutting trail • Links already there Person Company Person Person School ATTENDED ATTENDED W ORKED_FOR WORKED_FOR Degree C O M PLETED COM PLETED Priya Jennifer CMIS Skill Java Degree CS Dominic Work, Inc. Hire Me U HAS_SKILL H AS_SKILL HAS_SKILL COM PLETED
18. Beyond social networks Everywhere there are connections Supply chain Cybersecurity Routes, bottlenecks Access, activity Notes: bit.ly/jmhreifPearson
19. Beyond social networks Everywhere there are connections Recommendations Preferences, next thing Fraud detection Suspicious transactions, rings Notes: bit.ly/jmhreifPearson
20. Graphs add context and meaning Not just point-to-point, but HOW they connect Notes: bit.ly/jmhreifPearson
21. One model to rule them all… Conceptual = Logical = Physical data model Notes: bit.ly/jmhreifPearson
22. One model to rule them all… Conceptual = Logical = Physical data model Notes: bit.ly/jmhreifPearson
23. Cypher / GQL Graph query language • Functional and visual • Based on ASCII art • Declarative • Focus on patterns A B LIKES MATCH (A)-[:LIKES]->(B) Notes: bit.ly/jmhreifPearson
24. Cypher: write All about patterns TechCrunch Neo4j MENTIONS MERGE (:Article { site: ‘TechCrunch’}) -[:MENTIONS]-> (:Company { name: ‘Neo4j’}) NODE PROPERTY NODE PROPERTY LABEL LABEL Notes: bit.ly/jmhreifPearson
25. Cypher: read All about patterns TechCrunch Neo4j MENTIONS MATCH (:Article { site: ‘TechCrunch’} ) -[:MENTIONS]-> ( whom ) RETURN whom Notes: bit.ly/jmhreifPearson
26. A knowledge graph is a specific implementation of a graph database, where information is integrated from many different sources, representing the inherent knowledge of a particular domain. Notes: bit.ly/jmhreifPearson
27. Knowledge graphs Notes: bit.ly/jmhreifPearson
28. Exercise: Neo4j query in Java application Notes: bit.ly/jmhreifPearson
29. Layers of systems Today’s step-by-step Traditional systems neo4j Vector search / naive RAG GraphRAG Agents / MCP + neo4j Notes: bit.ly/jmhreifPearson
30. Q&A (5min) + Break (5min)
31. POLL! How are you using GenAI? Select all that apply. • At work for business • At school in classes or research • For personal life - organizing, planning, decision-making, help • For learning - whether business, school, or personal • Not using GenAI at all
32. Introduction to GenAI and GraphRAG
33. What is GenAI? Generative Arti fi cial Intelligence • Generating data in response to prompts • Learns patterns from training data • Generates new data with similar characteristics • Natural language response • Use cases: professional, personal, everything! Notes: bit.ly/jmhreifPearson
34. Demo: Large Language Model Notes: bit.ly/jmhreifPearson
35. How do hallucinations happen? LLM limitations • Lacking recent data • Too broad / general search • Not enough context • Uncertainty Notes: bit.ly/jmhreifPearson
36. Adding context to the LLM What does it do? • Guides it • Focuses / narrows search area • Adds to LLM knowledge • Reduces margin of error Notes: bit.ly/jmhreifPearson
37. RAG Pull data from external source • Retrieval • Data retrieved from external source • Augmented • Augments response with facts • Generation • Response in natural language Notes: bit.ly/jmhreifPearson
38. Types of context Search beyond keyword? • Databases (all kinds) • Documents/folders • Media (music/images/video) • Charts (reports/visuals) Notes: bit.ly/jmhreifPearson
39. Embeddings / Vectors • Data -> array of fl oating point numbers • Captures meaning and how it’s used • Comparable format! Convert data to a point in space Notes: bit.ly/jmhreifPearson
40. Vectors in the physical realm https://www.mathsisfun.com/algebra/vectors.html Notes: bit.ly/jmhreifPearson
41. Vectors in the technical realm Kings and Queens king − man + woman ≈ queen king man wom an 1 king man wom an 2 queen? 3 Notes: bit.ly/jmhreifPearson
42. Searching the data • Vectors -> semantic similarity • Proximity in vector space • Example: Library • Book classi fi cation - genre vs location of plot • More speci fi c + more relevant! Photo by Martin Adams on Unsplash …e ffi ciently Notes: bit.ly/jmhreifPearson
43. Vector search for RAG Notes: bit.ly/jmhreifPearson
44. Vector support in databases https://www.graft.com/blog/top-vector-databases-for-ai-projects
45. Exercise: Vector search with graph store Notes: bit.ly/jmhreifPearson
46. Benefits With RAG + LLM • Guide probability with context (grounding) • Focuses / narrows probability area • Adds to LLM knowledge • Reduce margin of error • How did the LLM get this answer? Photo by No Revisions on Unsplash Notes: bit.ly/jmhreifPearson
47. Where do vectors fall flat? How do you… • Limited metadata / connections • Verify vector representations? • Explain how it got to answer? • Similar not always highest relevance? Notes: bit.ly/jmhreifPearson
48. Layers of systems Today’s step-by-step Traditional systems neo4j Vector search / naive RAG GraphRAG Agents / MCP + neo4j + neo4j Notes: bit.ly/jmhreifPearson
49. Graphs -> how/why connections Analyze data from relationships • Relationships + entities • Designed for storing / navigating networks • Traceable • Explainable Notes: bit.ly/jmhreifPearson
50. Graphs connect the dots Structured + Unstructured • Flexible schema = natural fi t • Document connections, paths, etc • Connect varying data structures Structured Unstructured Chunk Article Organization Industry Person City Country Notes: bit.ly/jmhreifPearson
51. GraphRAG Graph as data source • Accuracy: extra context / related connections • Veri fi ability: check against understandable format • Explainability: trace path through graph for answer Notes: bit.ly/jmhreifPearson
52. Q&A (5min) + Break (5min)
53. Building solutions with GraphRAG
54. GraphRAG pattern
55. GraphRAG Patterns
56. Exercise: Build a GraphRAG solution Notes: bit.ly/jmhreifPearson
57. Considerations • Frameworks • Data extraction - unstructured vs structured • Retrieval strategies • Testing models, prompts, solutions, evals, etc Notes: bit.ly/jmhreifPearson
58. Framework choices 1.0 and beyond! • Spring AI (project within Spring framework) • Langchain4j (project supported by community+vendors) • MCP can integrate solutions across frameworks
59. “Unstructured” data Is it though? • Might be unstructured in appearance, length, styling • Structure is there! • Less obvious + consistent • Structure: • Themes • Moments • Stories
60. Graph retrieval example Superhero story/movie/comic/etc • Text = “Superhero saves citizens and defeats villain” • Search = “Which villains does Captain America defeat?” • Vector retrieval: • Any superheroes defeating villain (similarity) • Non-comic related examples of villains • Graph retrieval: • Superhero:Captain America - DEFEATS -> Villain Notes: bit.ly/jmhreifPearson
61. Hybrid retrieval example Superhero story/movie/comic/etc • Text = “Superhero saves citizens and defeats villain” • Search = “Which superheroes defeat villains?” • Vector retrieval: • Generic -> speci fi c • Graph retrieval: • Additional context on superhero details, networks, etc Notes: bit.ly/jmhreifPearson
62. Align retrieval strategy Vector, graph, hybrid, agents? • Determine questions • Match retrieval strategy • Iterate! Notes: bit.ly/jmhreifPearson
63. Layers of systems Today’s step-by-step Traditional systems neo4j Vector search / naive RAG GraphRAG Agents / MCP + neo4j + neo4j + neo4j Notes: bit.ly/jmhreifPearson
64. Agents and MCP
65. What is an agent? Software system using AI • What it does: • Evaluate -> sense environment, make decisions, take actions • Respond -> break down tasks/goal, execute tools, adapt to feedback • Work fl ows -> Autonomous system Notes: bit.ly/jmhreifPearson
66. What is MCP? Open standard • AI interact with sources, tools, and services • Modular, consistent access • Metaphors: • Universal adapter for AI • MCP server: microservice • Tool options: kids’ snacks • Vendor MCP servers: Docker containers Notes: bit.ly/jmhreifPearson
67. Demo: Agents and MCP Notes: bit.ly/jmhreifPearson
68. POLL! Which tech from this course are you most excited about? Select one. • Graph databases • LLMs and GenAI • RAG and/or GraphRAG • Agents • MCP • Something else • All the above
69. Q&A (10min)
70. Resources • Code: github.com/JMHReif/graphrag-fundamentals • GraphAcademy LLM courses: graphacademy.neo4j.com/knowledge-graph-rag • NODES 2025: neo4j.com/nodes • Knowledge graph ebook: dev.neo4j.com/jmhreif-kg-ebook Jennifer Reif jennifer.reif@neo4j.com @JMHReif github.com/JMHReif jmhreif.com linkedin.com/in/jmhreif Notes: bit.ly/jmhreifPearson