The Resource Description Framework (RDF) is a powerful formal knowledge representation language and a fundamental standard of the Semantic Web. It has its own vocabulary that defines core concepts and relations (e.g., rdf:type corresponds to the isA relationship), and a data model that enables machine-interpretable statements in the form of subject-predicate-object (resource-property-value) triples, called RDF triples, such as picture-depicts-book. The extension of the RDF vocabulary with concepts required to create controlled vocabularies and basic ontologies is called RDF Schema or RDF Vocabulary Description Language (RDFS). RDFS makes it possible to write statements about classes and resources, and express taxonomical structures, such as via superclass-subclass relationships.

Complex knowledge domains require more capabilities than what is available in RDFS, which led to the introduction of OWL. OWL supports relationships between classes (union, intersection, disjointness, equivalence), property cardinality constraints (minimum, maximum, exact number, e.g., every person has exactly one father), rich typing of properties, characteristics of properties and special properties (transitive, symmetric, functional, inverse functional, e.g., A ex:hasAncestor B and B ex:hasAncestor C implies that A ex:hasAncestor C), specifying that a given property is a unique key for instances of a particular class, and domain and range restrictions for properties.

一图胜千言!下面这张图应该可以加强克里斯托弗·古特里奇的观点 在这个回答中说语义网是一个“分层架构”。

来源:https://www.obitko.com/tutorials/ontologies-semantic-web/semantic-web-architecture.html

RDF、RDFS和OWL是表示日益复杂的信息或知识的方法。它们都可以用RDF/XML语法(或任何其他RDF序列化语法，例如Turtle或N3)序列化。

这些技术是相关的，应该是可互操作的，但它们有不同的起源，这可能就是为什么它们之间的关系很难掌握的原因。选择一种还是另一种取决于建模所需要的情况有多复杂。

表达能力的总结

RDF:简单的表示，主要关注实例及其类型的映射(RDF: type)。可以定义自定义属性来链接数据和创建三元组。使用SPARQL查询RDF数据。 在Turtle中序列化RDF的例子:

@prefix : <http://www.example.org/> .
:john    rdf:type           :Man .
:john    :livesIn  "New-York" .
:livesIn    rdf:type    rdf:Property .

RDFS: Some situations are not easily modelled by RDF alone, it is sometimes interesting to represent more complex relations like subclasses (the type of a type) for example. RDFS provides special means to represent such cases, with constructs like rdfs:subClassOf, rdfs:range or rdfs:domain. Ideally, a reasoner can understand the RDFS semantics and expand the number of triples based on the relations: For instance if you have the triples John a Man and Man rdfs:subClassOf Human then you should generate as well the triple John a Human. Note that this is not possible to do with RDF alone. RDFS data are queried using SPARQL. Example of RDFS serialised in Turtle:

@prefix : <http://www.example.org/> .
:john    rdf:type           :Man .
:Man    rdfs:subClassOf    :Human .
:john    :livesIn  "New-York" .
:livesIn    rdf:type    rdf:Property .
# After reasoning
:john    rdf:type    :Human .

OWL: The highest level of expressivity. Relation between classes can be formally modelled based on description logics (mathematical theory). OWL relies heavily on the reasoner, it is possible to express complex constructs such as chained properties for instance or restriction between classes. OWL serves to build ontologies or schema on the top of RDF datasets. As OWL can be serialised as RDF/XML, it is theoretically possible to query it via SPARQL, yet it is much more intuitive to query an OWL ontology with a DL query (which is usually a standard OWL class expression). Example of OWL constructs serialised in Turtle.

@prefix : <http://www.example.org/> .
:livesIn    rdf:type    owl:DatatypeProperty .
:Human    rdf:type    owl:Class .
:Man   rdf:type    owl:Class .
:Man    rdfs:subClassOf    :Human .
:John    rdf:type    :Man . 
:John    rdf:type    owl:NamedIndividual .

简而言之:

RDF定义了如何写东西 OWL定义了要写什么

正如前面的帖子所写的，RDF是一种规范，它告诉您如何定义三元组。

问题是RDF允许你定义任何东西，所以你可以像这样组成一个声明:

| subject | predicate | object |
|---------|-----------|--------|
| Alex    | Eats      | Apples |
| Apples  | Eats      | Apples |
| Apples  | Apples    | Apples |

这些三元组形成有效的RDF文档。

但是，从语义上讲，您知道这些语句是不正确的，RDF不能帮助您验证所编写的内容。

这不是一个有效的本体。

OWL规范准确地定义了您可以用RDF编写什么以获得有效的本体。

本体可以有几个属性。

这就是为什么OWL(版本1)定义了几个版本，如OWL DL, OWL Lite, OWL Full。

The Resource Description Framework (RDF) is a powerful formal knowledge representation language and a fundamental standard of the Semantic Web. It has its own vocabulary that defines core concepts and relations (e.g., rdf:type corresponds to the isA relationship), and a data model that enables machine-interpretable statements in the form of subject-predicate-object (resource-property-value) triples, called RDF triples, such as picture-depicts-book. The extension of the RDF vocabulary with concepts required to create controlled vocabularies and basic ontologies is called RDF Schema or RDF Vocabulary Description Language (RDFS). RDFS makes it possible to write statements about classes and resources, and express taxonomical structures, such as via superclass-subclass relationships.

Complex knowledge domains require more capabilities than what is available in RDFS, which led to the introduction of OWL. OWL supports relationships between classes (union, intersection, disjointness, equivalence), property cardinality constraints (minimum, maximum, exact number, e.g., every person has exactly one father), rich typing of properties, characteristics of properties and special properties (transitive, symmetric, functional, inverse functional, e.g., A ex:hasAncestor B and B ex:hasAncestor C implies that A ex:hasAncestor C), specifying that a given property is a unique key for instances of a particular class, and domain and range restrictions for properties.

我试图掌握语义网的概念。我发现这很难 来理解RDF和OWL之间的确切区别。是 OWL是RDF的扩展或者这两者是完全不同的 技术吗?

简而言之，是的，您可以说OWL是RDF的扩展。

In more detail, with RDF you can describe a directed graph by defining subject-predicate-object triples. The subject and the object are the nodes, the predicate is the edge, or by other words, the predicate describes the relation between the subject and the object. For example :Tolkien :wrote :LordOfTheRings or :LordOfTheRings :author :Tolkien, etc... Linked data systems use these triples to describe knowledge graphs, and they provide ways to store them, query them. Now these are huge systems, but you can use RDF by smaller projects. Every application has a domain specific language (or by DDD terms ubiquitous language). You can describe that language in your ontology/vocabulary, so you can describe the domain model of your application with a graph, which you can visualize show it to business ppl, talk about business decisions based on the model, and build the application on top of that. You can bind the vocab of your application to the data it returns and to a vocabulary known by the search engines, like microdata (for example you can use HTML with RDFA to do this), and so search engines can find your applications easily, because the knowledge about what it does will be machine processable. This is how semantic web works. (At least this is how I imagine it.)

Now to describe object oriented applications you need types, classes, properties, instances, etc... With RDF you can describe only objects. RDFS (RDF schema) helps you to describe classes, inheritance (based on objects ofc.), but it is too broad. To define constraints (for example one kid per chinese family) you need another vocab. OWL (web ontology language) does this job. OWL is an ontology which you can use to describe web applications. It integrates the XSD simpleTypes. So RDF -> RDFS -> OWL -> MyWebApp is the order to describe your web application in a more and more specific way.