TL;DR

An enterprise-domain benchmark shows that adding an ontology-based knowledge graph layer over SQL triples GPT-4 question-answering accuracy from 16% to 54% in zero-shot settings.

Summary

Enterprise SQL databases are hard for LLMs to query accurately because schemas alone carry no business semantics. This benchmark uses an insurance-domain enterprise schema paired with an ontology-based knowledge graph to measure how much semantic context affects accuracy. Testing GPT-4 with zero-shot prompting, direct SQL achieves only 16% accuracy while posing the same questions over a KG representation reaches 54%, demonstrating that semantic enrichment is a decisive factor independent of model capability.

Key contributions

  1. Introduces a benchmark using a real enterprise insurance schema to evaluate LLM question-answering accuracy.
  2. Demonstrates a threefold accuracy improvement (16% → 54%) by adding a KG layer over the same raw SQL data.
  3. Shows that ontology and R2RML mappings alone — without fine-tuning — supply sufficient semantic context for GPT-4.
  4. Establishes that popular text-to-SQL benchmarks underrepresent real enterprise complexity.

When to cite

  1. When arguing that knowledge graph enrichment is more impactful than model upgrades for enterprise Q&A accuracy.
  2. When supporting claims that schema structure alone is insufficient context for LLM-based SQL querying.
  3. When motivating the use of ontology-to-SQL mappings (OBDA/R2RML) as a production engineering asset.
  4. When benchmarking LLM readiness on domain-specific enterprise data with realistic query complexity.

LLM Knowledge Graph Prompt Engineering RAG Knowledge Graph Prompting for Multi-Document Question Answering Unifying Large Language Models and Knowledge Graphs - A Roadmap