Knowledge Graph
A data model that represents entities and their relationships in a graph structure. It is used to improve the accuracy of RAG and AI search.
## What is a Knowledge Graph? A knowledge graph is a data model that represents real-world entities (people, organizations, concepts, products, etc.) and the relationships between them using a graph structure consisting of nodes and edges. ### Differences from Relational Databases While relational databases manage data using rows and columns in tables, knowledge graphs store data as subject → predicate → object triples. Like "Tokyo → is the capital of → Japan," it is a structure in which the data itself carries meaning. This structure allows multi-hop relationship queries such as "Who are the competitors of Company A's business partners?" to be naturally handled through graph traversal — the kind of query that would require joining multiple tables in a relational database. ### Integration with AI Search Knowledge graphs are a core technology in GraphRAG. Whereas vector search retrieves documents based on semantic similarity of text, knowledge graphs retrieve structured knowledge by traversing paths between entities. Combining both approaches is expected to reduce hallucinations and improve the accuracy of multi-hop reasoning. ### Construction Costs The biggest hurdle with knowledge graphs is the cost of building and maintaining them. Although automation has advanced through LLM-based entity extraction and relationship inference, there remain situations where expert review is indispensable for accurately mapping domain-specific knowledge systems into a graph.
Related Terms
Agentic RAG
Agentic RAG is an architecture in which an LLM autonomously and iteratively generates search queries, evaluates results, and decides whether to re-retrieve information as an agent, achieving answer accuracy that cannot be obtained with simple single-turn RAG.
RAG (Retrieval-Augmented Generation)
RAG (Retrieval-Augmented Generation) is a technique that improves the accuracy and currency of responses by retrieving relevant information from external knowledge sources and appending the results to the input of an LLM.
Gherkin notation
Gherkin notation is a structured format for describing software behavior in natural language using three steps: Given (precondition), When (action), and Then (outcome). It is widely used as the standard notation for .feature files read by the test automation tool Cucumber.
