Overview
Knowledge graphs represent information as a network of 'entities' (nodes) and 'relationships' (edges). For example, 'Albert Einstein' (node) 'was born in' (edge) 'Ulm' (node).
Key Features
- Semantic Meaning: Relationships have specific types and meanings.
- Interconnectivity: Allows for discovering complex paths and connections between disparate data points.
- Reasoning: Can be used to infer new facts that aren't explicitly stated.
Use in AI
Knowledge graphs are often used to provide 'ground truth' or factual context to AI systems, helping to reduce hallucinations in LLMs and power more accurate search results (e.g., Google's Knowledge Panel).