COMPEL Glossary / knowledge-graph
Knowledge Graph
A knowledge graph is a structured representation of real-world entities (people, places, concepts, products) and their relationships, stored in a graph database that enables sophisticated querying and reasoning.
What this means in practice
Knowledge graphs help AI systems understand connections between concepts: 'Product A is manufactured in Factory B, which is in Region C, which is subject to Regulation D.' This relational understanding is used in search engines (Google's Knowledge Graph powers many search results), recommendation systems, fraud detection (identifying suspicious relationship networks), and question-answering applications. For enterprise AI, knowledge graphs can enhance LLM capabilities by providing structured factual context that reduces hallucination risk. Knowledge graphs are part of the emerging data infrastructure landscape assessed in COMPEL Domain 10.
Why it matters
Knowledge graphs help AI systems understand relationships between real-world entities, enabling sophisticated capabilities in search, recommendations, fraud detection, and question answering. For enterprise AI, knowledge graphs can enhance LLM capabilities by providing structured factual context that reduces hallucination risk. Organizations that build domain-specific knowledge graphs create a reusable asset that improves the quality and reliability of multiple AI applications.
How COMPEL uses it
Knowledge graphs are part of the emerging data infrastructure landscape assessed in Domain 10 (Data Infrastructure) during Calibrate. During Model, knowledge graph architecture decisions are made when designing enterprise search, RAG systems, or domain-specific reasoning capabilities. The Technology pillar design in Module 3.3 covers knowledge graph integration patterns. The Evaluate stage measures the impact of knowledge graphs on AI system accuracy and hallucination rates.
Related Terms
Other glossary terms mentioned in this entry's definition and context.