Semantic Web

By /

The Semantic Web improves the World Wide Web by adding standards set by the World Wide Web Consortium. Its an idea thought of by Tim Berners-Lee, and its purpose is to make online data readable to machines. (Source)

What is Semantic Web?

Semantic Web was originally made for these three things

  • Automation of Information retrieval
  • The internet of things
  • Personal Assistants

Over the years, it evolved into two important data types:

Linked Open Data

Linked Open Data is a way of publishing structured data so that it can be connected across different websites. Data is connected using Uniform Resource Identifiers (URIs) which means that every resource gets a unique HTTP web address. Useful data must be provided in the URI using the standards (RDF, SPARQL).

What Linked Open Data Includes:

Factual data on entities or topics

  • For example a movie, a company or a thing

Ontologies define

  • Objects: Person, Organization, Location, Document
  • Relations: Parent of, made by
  • Attributes: Birthdate, population size

Semantic Metadata

Semantic Metadata is semantic tags you stick on regular web pages to say what they’re really about. This makes it easier to find stuff when you search and prevents confusion for the web. A website about music instruments could be tagged with keywords like, guitar, drums, bass or piano.

If you want a solid link between what the page is about and the metadata, use structured formats in for the form of schema. schemawriter produce webpage schema, and most other schema types, read more about webpage schema here.

Schema.org is the leading one – founded by Google, Yahoo, Microsoft, and Yandex. In 2015, a study from the University of Mannheim said 30% of web pages had semantic metadata. (Source)

Standardization

Semantic Web standards are handled by W3C.

Web 3.0 builds on them.

Core Technologies

  • The Semantic Web uses formats and protocols.
  • They define concepts, terms, and links between data.
  • They are W3C standards:
  • RDF – basic model for describing web resources.
  • RDFS – adds structure with classes and properties.
  • OWL – adds logic: relations, constraints, equivalence.
  • SPARQL – query language for RDF data.
  • SKOS – framework for taxonomies and vocabularies.
  • N-Triples – simple, line-based RDF syntax.
  • Turtle – compact, readable RDF format.
  • Notation3 (N3) – like Turtle, but supports logic.
  • RIF – rule exchange across systems, with multiple dialects.
  • JSON-LD – linked data in JSON format.
  • ActivityPub – protocol for server-to-server and client-to-server interaction (used by Mastodon).

Stack and Function

The Semantic Web has layers.

Each layer adds function.

  • XML – syntax for structure. No meaning.
  • XML Schema – constraints for XML content.
  • RDF – expresses data models and relationships.
  • RDFS – describes hierarchies and types in RDF.
  • OWL – adds expressiveness: logic, cardinality, class relations.
  • SPARQL – extracts and manipulates RDF data.
  • RIF – defines executable rule logic. Uses XML. Has variants (BLD, PRD).
  • Turtle is commonly used.
  • XML is less important now.

Standardization Status

Established standards:

  • RDF
  • RDFS
  • OWL
  • SPARQL
  • RIF
  • URI
  • Unicode
  • XML (Source)

Semantic Web and SEO

In SEO, Semantic Web is used a lot. Search engines like Google use structured data to understand what a page is about. Structured data turns static pages into part of a broader semantic network. It links content to related entities across the web. The better Google understands a page the better are the odds of the page ranking. This is done through schemas for common products, books or events. Adding this data to a website helps search engines connect the subject of the page better than just matching keywords. (Source)

We made an SEO experiment, to see if schema markup from schema.org improved Google rankings. The short answer is yes, it does. The results showed that in only 5 days, we ranked number one on Google for the target keyword “automated webpage schema”. We reached number one by implementing automated webpage schema using JSON-LD and adding entity-optimized content from schemawriter.ai. You can read about the full experiment here: Does schema improve Google rankings?

Scroll to Top