RDF Schema (RDFS)

The rdfs:Class is used to define categories or types of resources. It is the RDFS construct for declaring a class, analogous to a type or category in object-oriented programming or a table in a database.

• Declaration: A class is declared using rdf:type rdfs:Class. For example, :Person rdf:type rdfs:Class.
• Instantiation: An individual resource becomes an instance of a class using the rdf:type property. For example, :Alice rdf:type :Person.
• Core Purpose: Classes provide the primary mechanism for organizing and categorizing instances within an RDF graph, forming the backbone of a simple taxonomy.

The rdfs:subClassOf property establishes a hierarchical relationship between two classes, indicating that all instances of the subclass are also instances of the superclass. This enables inheritance of characteristics.

• Semantics: If A rdfs:subClassOf B, then any resource that is an rdf:type of A is inferred to also be an rdf:type of B.
• Example: :Employee rdfs:subClassOf :Person. Any individual typed as an :Employee is automatically inferred to be a :Person.
• Transitivity: The property is transitive. If :SoftwareEngineer rdfs:subClassOf :Employee and :Employee rdfs:subClassOf :Person, then :SoftwareEngineer is implicitly a subclass of :Person.

The rdf:Property class is the type for all RDF predicates. In RDFS, properties are first-class citizens that can be described and organized.

• Declaration: A property is declared as an instance of rdf:Property. For example, :hasName rdf:type rdf:Property.
• Domain & Range: RDFS allows you to constrain a property's usage with rdfs:domain and rdfs:range.
  • rdfs:domain specifies the class of the subject. :hasName rdfs:domain :Person means any subject using :hasName is inferred to be a :Person.
  • rdfs:range specifies the class (or datatype) of the object. :hasName rdfs:range xsd:string means the object of :hasName is a string literal.

The rdfs:subPropertyOf property creates a hierarchy between RDF properties, allowing for the specialization of relationships.

• Semantics: If P1 rdfs:subPropertyOf P2, then any triple X P1 Y implies the triple X P2 Y.
• Example: :hasMother rdfs:subPropertyOf :hasParent. If :Alice :hasMother :Carol, it can be inferred that :Alice :hasParent :Carol.
• Use Case: This enables querying at different levels of granularity. A query for all :hasParent relationships will also retrieve those made with the more specific :hasMother property.

RDFS provides vocabulary for describing groups of resources, extending the basic RDF container model (rdf:Bag, rdf:Seq, rdf:Alt).

• rdfs:Container: A superclass of the RDF container classes.
• rdfs:member: A generic property to relate a container to its members, subsuming rdf:_1, rdf:_2, etc.
• rdfs:ContainerMembershipProperty: The class of properties rdf:_1, rdf:_2, etc. This formalizes their status.
• Limitation: RDFS does not provide the rich, formal semantics for collections found in OWL (e.g., cardinality restrictions, disjointness). Its support is primarily for labeling and basic membership.

RDFS includes essential properties for labeling, documenting, and linking resources, which are critical for human understanding and interoperability.

• rdfs:label: Provides a human-readable name for a resource. It is language-tagged, e.g., :Person rdfs:label "Person"@en.
• rdfs:comment: Offers a longer description or explanation of a resource.
• rdfs:seeAlso: Indicates a resource that provides additional information about the subject.
• rdfs:isDefinedBy: Points to the resource defining the subject, often an ontology or schema document.
• rdfs:Literal: The class of literal values such as strings and numbers.
• rdfs:Datatype: The class of RDF datatypes, a subclass of rdfs:Class.

The foundational data model upon which RDFS is built. RDF represents information as a directed graph of subject-predicate-object triples. It provides the universal syntax for expressing statements about resources, forming the atomic unit of semantic data. RDFS extends this model by adding a vocabulary for organizing these triples into structured schemas.

• Triple: The basic statement: <subject> <predicate> <object>.
• Graph-Based: Data is a set of interconnected triples, not rows and columns.
• Universal Identifiers: Uses URIs to uniquely identify resources, enabling global data linking.

A more expressive ontology language that builds upon RDFS. While RDFS defines basic classes and hierarchies, OWL adds extensive capabilities for defining complex constraints and relationships using formal logic (description logics). It is used when precise, machine-interpretable semantics are required for automated reasoning.

• Greater Expressivity: Defines property characteristics (e.g., symmetry, transitivity), cardinality restrictions, and disjoint classes.
• Formal Semantics: Has a precise mathematical meaning, enabling logical inference.
• Three Profiles: OWL 2 EL, QL, and RL offer trade-offs between expressivity and computational complexity.

An RDF-based vocabulary for organizing knowledge like thesauri, taxonomies, and classification schemes. It operates at a conceptual level similar to RDFS but is optimized for human navigation and indexing rather than automated reasoning. SKOS concepts can be linked to RDFS/OWL classes for hybrid modeling.

• Core Relations: skos:broader, skos:narrower, skos:related for hierarchical and associative links.
• Lexical Labels: skos:prefLabel, skos:altLabel for human-readable names in multiple languages.
• Mapping: skos:exactMatch, skos:closeMatch for aligning concepts across different KOS.

A W3C standard for validating RDF graphs against a set of conditions. While RDFS and OWL use an open-world assumption for inference, SHACL uses a closed-world assumption to define the expected structure and data integrity rules of an RDF graph, similar to schema validation in relational databases.

• Data Integrity: Validates cardinality, data types, value ranges, and logical constraints.
• Shape-Based: Defines "shapes" that target nodes must conform to.
• Complementary Role: Often used alongside RDFS/OWL; OWL infers new facts, SHACL validates existing data.

The standard query language for RDF data. SPARQL is used to retrieve and manipulate information stored in RDF format, including data structured using RDFS vocabularies. It enables complex graph pattern matching, aggregation, and subqueries across distributed datasets.

• Graph Pattern Matching: The WHERE clause specifies patterns of triples to match in the graph.
• Inference Aware: Can be executed with a reasoner to query both explicit and RDFS/OWL-inferred triples.
• Protocol: Includes a standardized HTTP protocol (SPARQL Protocol) for remote query execution.

The family of formal knowledge representation languages that provide the logical underpinnings for OWL and, by extension, the more expressive aspects of semantic modeling beyond RDFS. Description Logics focus on defining concepts (classes), roles (properties), and individuals, and provide computable reasoning services.

• Logical Foundation: OWL is essentially a web syntax for specific Description Logics (e.g., SROIQ).
• Reasoning Services: Enables consistency checking, concept subsumption (classification), and instance checking.
• Computable Trade-offs: Different Description Logics offer varying balances of expressivity and computational complexity.