Structured Data
Knowledge Graph
A knowledge graph is a network of interconnected entities, such as people, places, and concepts, that describes real-world objects and the specific relationships between them in a machine-readable format.
// definition
A knowledge graph is a structured database that organizes information around distinct entities and the semantic connections between them rather than storing data in isolated tables. Nodes in the graph represent real-world entities, such as organizations, locations, or products, while the edges connecting the nodes express explicit relationships, such as location or ownership. These connections are typically expressed using semantic web standards, such as Resource Description Framework (RDF) or Web Ontology Language (OWL), which assign precise, machine-readable meanings to both entities and properties.
Unlike a traditional relational database that relies on rigid columns and tables, a knowledge graph enables systems to deduce new facts through context and reasoning. It is also distinct from a standard graph database, as a knowledge graph enforces an underlying schema or ontology that standardises real-world concepts across disparate datasets.
// why it matters
When search engines build internal knowledge graphs, they shift from matching literal text keywords to understanding user intent and entities. Publishing structured metadata on a website allows search crawlers to incorporate an organization's brand, products, and personnel directly into these knowledge networks. This integration directly influences how software systems interpret content relevance and can trigger prominent visual features in search interface results. As a result, websites that accurately define their entity relationships make their information easier for automated systems to extract, index, and surface for complex conversational user queries.
// example
Consider an online bookstore page for a biography. The page contains structured code declaring that the subject is an author entity named Jane Doe, who wrote a specific book entity published by a publishing company entity. Instead of recognizing only the word string Jane Doe, a search engine processes these interconnected data points into its knowledge graph, linking the author to her published works, awards, and birthplace.
Questions and Answers
- What is the difference between a knowledge graph and a database?
- A knowledge graph differs from a traditional database by focusing primarily on the semantic relationships between data points rather than storing entries in rigid rows and columns. While standard databases require strict schemas to query tables, a knowledge graph uses an interconnected web of real-world entities, allowing software systems to infer context, connect disparate sources, and answer complex queries more effectively.
- How do search engines build their own knowledge graphs?
- Search engines construct knowledge graphs by crawling public web pages, extracting structured markup, and combining information from verified public datasets. They use natural language processing to identify real-world entities within unstructured text and analyze how these entities relate to one another. Over time, automated systems validate and refine these connections to build a comprehensive map of real-world information.
- How can a business get added to a knowledge graph?
- A business enters a search engine's knowledge graph by consistently publishing structured data markup across its official website. Maintaining consistent details across authoritative third-party directories, public databases, and official social profiles also helps automated algorithms recognize the business as a distinct entity and map its relationships to products, executives, and geographic locations accurately.