Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Semantic Web: Past, Present, and Future (with Machine Learning on Knowledge Graphs and Language Models on Knowledge Graphs)

Ansgar Scherp, Gerd Groener, Petr Škoda, Katja Hose, Maria-Esther Vidal

TL;DR

This paper surveys the Semantic Web from its foundational concepts (RDF, RDFS, OWL, SPARQL, Linked Data) to modern developments, including machine learning on knowledge graphs and language models operating on graph-structured data. It analyzes representation, creation, validation, reasoning, querying, and trust/provenance, and then synthesizes advances in graph embeddings, graph neural networks, graph-aware language models, and retrieval-augmented generation. The authors discuss practical applications (Schema.org, Wikidata, semantic search) and industry impact, while outlining challenges in data quality, provenance, and scalable reasoning. The work concludes with a forward-looking outlook on neuro-symbolic AI, natural-language interfaces, and evolving trust and crypto mechanisms as key enablers for the next Semantic Web era.

Abstract

Ever since the vision was formulated, the Semantic Web has inspired many generations of innovations. Semantic technologies have been used to share vast amounts of information on the Web, enhance them with semantics to give them meaning, and enable inference and reasoning on them. Throughout the years, semantic technologies, and in particular knowledge graphs, have been used in search engines, data integration, enterprise settings, and machine learning. In this paper, we recap the classical concepts and foundations of the Semantic Web as well as modern and recent concepts and applications, building upon these foundations. The classical topics we cover include knowledge representation, creating and validating knowledge on the Web, reasoning and linking, and distributed querying. We enhance this classical view of the so-called ``Semantic Web Layer Cake'' with an update of recent concepts. These include provenance, security and trust, as well as a discussion of practical impacts from industry-led contributions. We also provide an overiew of shallow and deep machine learning methods for knowledge graphs and discuss the relation of language models and knowledge graphs. We conclude with an outlook on the future directions of the Semantic Web.

Semantic Web: Past, Present, and Future (with Machine Learning on Knowledge Graphs and Language Models on Knowledge Graphs)

TL;DR

This paper surveys the Semantic Web from its foundational concepts (RDF, RDFS, OWL, SPARQL, Linked Data) to modern developments, including machine learning on knowledge graphs and language models operating on graph-structured data. It analyzes representation, creation, validation, reasoning, querying, and trust/provenance, and then synthesizes advances in graph embeddings, graph neural networks, graph-aware language models, and retrieval-augmented generation. The authors discuss practical applications (Schema.org, Wikidata, semantic search) and industry impact, while outlining challenges in data quality, provenance, and scalable reasoning. The work concludes with a forward-looking outlook on neuro-symbolic AI, natural-language interfaces, and evolving trust and crypto mechanisms as key enablers for the next Semantic Web era.

Abstract

Ever since the vision was formulated, the Semantic Web has inspired many generations of innovations. Semantic technologies have been used to share vast amounts of information on the Web, enhance them with semantics to give them meaning, and enable inference and reasoning on them. Throughout the years, semantic technologies, and in particular knowledge graphs, have been used in search engines, data integration, enterprise settings, and machine learning. In this paper, we recap the classical concepts and foundations of the Semantic Web as well as modern and recent concepts and applications, building upon these foundations. The classical topics we cover include knowledge representation, creating and validating knowledge on the Web, reasoning and linking, and distributed querying. We enhance this classical view of the so-called ``Semantic Web Layer Cake'' with an update of recent concepts. These include provenance, security and trust, as well as a discussion of practical impacts from industry-led contributions. We also provide an overiew of shallow and deep machine learning methods for knowledge graphs and discuss the relation of language models and knowledge graphs. We conclude with an outlook on the future directions of the Semantic Web.

Paper Structure

This paper contains 39 sections, 10 figures.

Table of Contents

  1. Introduction
  2. Motivating Example
  3. Architecture of the Semantic Web
  4. Representation of Graph Data
  5. Linked Graph Data on the Web
  6. Ontologies
  7. Types and Examples of Ontologies
  8. Distributed Network of Ontologies and Ontology Patterns
  9. Creation and Validation of Graph Data
  10. Graph Data Creation
  11. Quality and Validation of Graph Data
  12. Reasoning over and Linking of Graph Data
  13. Reasoning over Graph Data
  14. Linking of Objects and Concepts
  15. Querying of Linked Data
  16. ...and 24 more sections

Figures (10)

  • Figure 1: Example of the BBC ontology with links to other ontologies (notation based on UML, here without prefix for namespaces).
  • Figure 2: Representation of the components of the so-called "Semantic Web Layer Cake". W3C language standards are shown in dark gray. Current developments are shown in light gray.
  • Figure 3: Linked Data example for ABBA.
  • Figure 4: Ontology layers combining the foundational ontology DOLCE, the multimedia metadata ontology M3O, domain-specific extensions to M3O for annotating audio data and music, and a domain ontology for albums and tracks.
  • Figure 5: Structure of a relational data mapping (source: W3CR2RML).
  • ...and 5 more figures