NODES 2024 - LLM Query Benchmarks: Cypher vs SQL

Modern businesses have lots of diverse data for making decisions. However, the complexity of analysing this data can make it practically useless for decision-makers. How can we make querying complex data more accessible? This talk explores the cutting-edge capabilities of large language models (LLMs) in translating user requests into Cypher and SQL. We assess such factors as dataset complexity, query intricacy, and the impact of few-shot examples on query quality. Which LLM does a better translator job? Elena Kohlwey and Rinat Abdullin will present cases from multiple industries and introduce an LLM benchmark to pick the best language model for your needs. Get certified with GraphAcademy: https://dev.neo4j.com/learngraph Neo4j AuraDB https://dev.neo4j.com/auradb Knowledge Graph Builder https://dev.neo4j.com/KGBuilder Neo4j GenAI https://dev.neo4j.com/graphrag