Ontology

The fundamental categories or types of things within a domain. Classes define the intensional definition of a group of entities.

• Formal Definition: A set of individuals sharing common characteristics.
• Example: In a financial ontology, FinancialInstrument is a class; Equity and Bond are its subclasses.
• Mechanism: Classes are arranged in a taxonomic hierarchy (subclass/superclass) supporting inheritance of properties.

The concrete, specific objects or ground-level entities that are members of a class. They represent the extensional definition or the actual data.

• Formal Definition: A single, unique object in the domain of discourse.
• Example: AAPL (Apple Inc.) is an individual of the class PublicCompany.
• Distinction: Unlike classes, individuals do not have sub-instances. They are the leaf nodes of the ontology.

Binary relations that define the attributes of classes or the relationships between individuals. Properties are first-class citizens in an ontology.

• Object Properties: Link an individual to another individual (e.g., hasCEO links Apple to Tim Cook).
• Datatype Properties: Link an individual to a literal value (e.g., hasRevenue links Apple to "383.29B"^^xsd:decimal).
• Annotation Properties: Add metadata like rdfs:label or rdfs:comment.

Assertions that constrain the interpretation of the ontology and enable logical inference. Axioms transform a simple data model into a reasoning engine.

• Domain/Range Restrictions: Specify that the property hasCEO only applies to Organization and points to Person.
• Cardinality Constraints: Assert that a Person has exactly one dateOfBirth.
• Disjointness: Declare that Man and Woman are disjoint classes, preventing logical contradictions.

The semantic links that define how entities interact. Unlike a simple database foreign key, relations carry explicit formal semantics.

• Transitive: If Paris isPartOf France, and France isPartOf Europe, then Paris isPartOf Europe.
• Symmetric: If Bob isSpouseOf Alice, then Alice isSpouseOf Bob.
• Inverse: hasParent is the inverse of hasChild.

Anonymous class expressions that define a class by constraining its properties. These are the building blocks of description logic.

• Universal Restriction (allValuesFrom): A HappyPerson has all children HappyPerson.
• Existential Restriction (someValuesFrom): A Parent has at least one child Person.
• HasValue: A RedWine has color Red. This allows classification of individuals based on their property values.

A hierarchical classification scheme that organizes concepts into parent-child relationships using broader/narrower term associations. Unlike an ontology, a taxonomy defines only is-a relationships (e.g., 'a dog is-a mammal') without specifying properties, constraints, or lateral relationships. Taxonomies provide the controlled vocabulary backbone upon which richer ontological relationships are built.

A predefined, restricted list of authorized terms used for indexing and retrieval to eliminate ambiguity. Controlled vocabularies enforce terminological consistency by mandating that only one term represents a concept (e.g., 'automobile' not 'car'). They are the atomic building blocks of taxonomies and ontologies, ensuring that semantic reasoning operates on unambiguous, normalized labels.

A structured data model that instantiates an ontology by representing specific entities as nodes and their relationships as edges. While an ontology defines the schema (classes and properties), a knowledge graph populates it with real-world instances (e.g., 'Elon Musk' is an instance of the class 'Person' with property 'founded' linking to 'SpaceX'). Search engines use knowledge graphs to store deterministic facts for entity disambiguation.

The process of matching a named entity reference in a dataset to a unique, canonical identifier in a knowledge base (e.g., a Wikidata Q-ID). Reconciliation resolves ambiguity by aligning local data records with a shared ontological framework, enabling interoperability between disparate systems and consolidating authority signals for search engines.

A collaborative, community-driven vocabulary of structured data schemas co-founded by Google, Microsoft, Yahoo, and Yandex. Schema.org provides a lightweight, web-native ontology for defining entities like Organization, Product, and Event using JSON-LD. It is the de facto standard for communicating ontological relationships to search engines to power rich results and knowledge graph panels.

A formal semantic web standard from the W3C for defining complex ontologies with rich axioms. OWL extends RDF Schema with constructs like cardinality constraints, disjointness, and inverse properties, enabling logical inference engines to derive new facts from explicitly stated ones. It is the foundation for heavyweight enterprise ontologies requiring automated reasoning.