Ontologies and knowledge graphs are important approaches to organizing structured knowledge in artificial intelligence. An ontology defines the concepts that exist in a particular domain, the categories they belong to, their properties, and the relationships that can connect them. It may also specify constraints and logical rules that describe how concepts relate to one another. A knowledge graph represents information as entities connected by labeled relationships. A common representation is a subject-predicate-object triple, such as 'Dhaka — located in — Bangladesh'. Ontologies and knowledge graphs are related but not identical. An ontology provides a conceptual model or vocabulary, while a knowledge graph contains represented entities and facts using relationships. A knowledge graph may use an ontology as its schema, although not every knowledge graph depends on a formal ontology. These techniques help systems integrate information from multiple sources, answer relationship-based queries, and discover connections between entities. Knowledge graphs are widely used in search engines, recommendation systems, digital assistants, and enterprise knowledge management.
Ontologies and Knowledge Graphs
Ontologies define concepts, categories, properties, and relationships within a domain, while knowledge graphs represent entities and their relationships as interconnected graph data.
Knowledge Graph Triple = (Subject, Predicate, Object)
Explanation
Variables
| Symbol | Meaning |
|---|---|
Subject | The entity or resource from which a relationship begins. |
Predicate | The named relationship connecting the subject to the object. |
Object | The entity or value connected to the subject by the predicate. |
Ontology | A formal model defining concepts, properties, relationships, and constraints in a domain. |
Entity | A real-world or abstract object represented in a knowledge graph. |
Example or application
A knowledge graph can represent the triple (Dhaka, located in, Bangladesh). An ontology for geography may define City and Country as concepts and located in as an allowed relationship between them.
Text
An ontology provides a structured vocabulary and formal description of a domain. A knowledge graph uses entities and relationships to connect information, often relying on an ontology or schema to define how that information is organized.
Key Points
An ontology defines a structured conceptual model for a domain.
A knowledge graph represents entities and their relationships.
Subject-predicate-object triples are a common graph representation.
Ontologies can provide schemas and constraints for knowledge graphs.
Knowledge graphs support information integration and relationship-based queries.
Applications
Search engine knowledge panels
Recommendation systems
Enterprise knowledge management
Biomedical information integration
Question-answering systems
Digital assistants
Learning Outcomes
Define an ontology and a knowledge graph.
Explain the difference between a conceptual schema and graph facts.
Interpret subject-predicate-object triples.
Describe how knowledge graphs connect information from multiple sources.
Practice Questions
What is an ontology?
What are the three parts of a knowledge graph triple?
How does an ontology differ from a knowledge graph?
Why are knowledge graphs useful for question answering?
Related Concepts
Semantic Networks
Knowledge Representation in AI
RDF
OWL
Graph Databases
Common Mistakes
Treating ontologies and knowledge graphs as exactly the same thing.
Confusing an entity with the relationship connecting two entities.
Assuming every knowledge graph must use a formal ontology.
Ignoring the meaning and constraints of relationship types.
Sources and verification
Sources are shown as provided in the project data. Not every source has yet been independently verified. Please report errors or missing citations.
- W3C. RDF 1.1 Concepts and Abstract Syntax.
- W3C. OWL 2 Web Ontology Language: Document Overview.
- Hogan, A. et al. Knowledge Graphs, ACM Computing Surveys, 2021.
Suggest a correction or source · More in Artificial Intelligence & Machine Learning