GraphRAG: Data with Context Jennifer Reif [email protected] @JMHReif github.com/JMHReif jmhreif.com linkedin.com/in/jmhreif Photo by Tomasz Frankowski on Unsplash 

Who is Jennifer Reif? Developer Advocate, Neo4j • Continuous learner • Technical speaker • Tech blogger, podcaster • Other: geek Jennifer Reif [email protected] @JMHReif github.com/JMHReif jmhreif.com linkedin.com/in/jmhreif 

What is a graph? Answers through relationships • How many coworkers shared classes/degrees? • What are common degree journeys? • How many alumni re- enroll for higher degrees? • Who else went to a school and works for company? Person Degree Degree Company Person Person School ATTENDED ATTENDED W ORKED_FOR WORKED_FOR EN RO LLED_IN ENROLLED_IN ENROLLED_IN Degree C O M PLETED COM PLETED Edward Jones Jacob Jennifer SIUE Music CMIS CS Adrian 

Graphs add context and meaning Not just point-to-point, but HOW they connect 

Data model 

One model to rule them all… Conceptual = Logical = Physical data model 

One model to rule them all… Conceptual = Logical = Physical data model 

Knowledge Graphs 

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. 

Layers of data value 

Demo! 

AI 

Layers of AI • More layers = better result • Complexity vs value 

Throw an LLM at it Doesn’t often work • LLM strengths: • human-consumable data • probabilistic answers • LLM weaknesses: • complex logic • recognizing nuances • Solo LLM = great -> integrated LLM = more great? 

RAG architecture • Retrieval • Data retrieved from external source • Augmented • Augments response with facts • Generation • Response in natural language Prompt + Relevant Information LLM API LLM
 Chat API User Database Search Prompt Response Relevant Results / Documents 2 3 1 Database 

LLMs take text, not databases And context window limit 

Lexical to Semantic search Semantic layer • Keyword searches aren’t enough • Data dictionaries are complex and limited • How do we search relevant/contextual info? 

Embeddings / Vectors Convert data to a point in space • Series of numbers • 100s or 1000s of dimensions • Dimension = interesting feature / characteristic 

Vectors in the physical realm https://www.mathsisfun.com/algebra/vectors.html 

Vectors in the technical realm Kings and Queens king − man + woman ≈ queen king man wom an 1 king man wom an 2 queen? 3 

LLMs take certain formats, not databases Vectors 

How do we search the vectors? Similarity search • Expensive queries (compare to every vector) • Approximate nearest neighbor (k-ANN) • Example: Library • Book classi fi cation - genre vs location of plot • Smaller search set = smaller retrieval time! Photo by Martin Adams on Unsplash 

Provide prompt + context to LLM Vectors Text 

Vector-only RAG Similarity is insu ff i cient for rich reasoning/explainability • Missing context, relationships • Vector similarity != relevance • Lack explainability 

GraphRAG layer Graph DB as a data source • More relevant data to provide as context • Quality connected data, higher accuracy • Explainability, veri fi ability 

Explainable AI Why graph with GenAI? • How did the LLM get this answer? • Graphs: • Incorporating siloed data into result • Add extra context/related info to graph • Better veri fi cation through understandable format Photo by No Revisions on Unsplash 

Knowledge graph as knowledge base Document in KG -> Knowledge Graph 

Explainability layer 

Nothing is a silver bullet LLM is (of sorts) mind of its own • Can’t guarantee a consistent answer • Prompt engineering • Context window limits 

Let’s code! 

No content

No content

Resources • Github repository (today’s code): github.com/JMHReif/ai-pet-travel • GraphAcademy LLM courses: graphacademy.neo4j.com/categories/llms/ • Docs for Spring AI: docs.spring.io/spring-ai/reference/api/vectordbs/neo4j.html • Knowledge graph ebook: https://neo4j.com/whitepapers/developers-guide-how- to-build-knowledge-graph/ Jennifer Reif [email protected] @JMHReif github.com/JMHReif jmhreif.com linkedin.com/in/jmhreif