Skip to content
Ontology Training · Knowledge Graphs

Give your data a meaning a machine can check.

Open courses in RDF, OWL, SPARQL and SHACL, plus applied tracks for construction, insurance, space and safety. The teaching sits on ontologies and crosswalks we publish, not on slides about someone else’s stack.

Start with Foundations → Explore All Tracks
  • 100% Open Access · No Payment Required
  • Interactive Quiz & Verification on Every Lesson
  • Grounded on Published Production Standards
🏢 Entity :InsurerOrg 🔗 Relation :hasLEI 🛡️ Shape :LEIShape ✓ Validated 100% Conforms
Node Inspector • Click any node above
:InsurerOrg a owl:Class ; rdfs:label "Authorised Insurer or Reinsurer" ; sh:property :LEIShape .
@prefix ex: <https://tesseract.academy/ontology/> .
@prefix owl: <http://www.w3.org/2002/07/owl#> .
@prefix sh:  <http://www.w3.org/ns/shacl#> .

ex:AuthorisedInsurer a owl:Class ;
    rdfs:label "Authorised European Insurer" ;
    rdfs:subClassOf ex:FinancialInstitution .

ex:Entity-84920 a ex:AuthorisedInsurer ;
    ex:legalEntityIdentifier "5493006MHB84DD0ZWV18" ;
    ex:registrationStatus "Active" ;
    ex:validatedTriples 51355 .
Knowledge Scale 51,355 Triples
Constraint Report 0 Violations 
ex:LEIShape a sh:NodeShape ;
    sh:targetClass ex:AuthorisedInsurer ;
    sh:property [
        sh:path ex:legalEntityIdentifier ;
        sh:minCount 1 ;
        sh:pattern "^[0-9A-Z]{18}[0-9]{2}$" ;
        sh:message "LEI must conform to ISO 17442 check digits." ;
    ] .
SELECT ?insurer ?lei ?status WHERE {
  ?insurer a ex:AuthorisedInsurer ;
           ex:legalEntityIdentifier ?lei ;
           ex:registrationStatus ?status .
  FILTER(?status = "Active")
} LIMIT 5
⚡ Query Executed • 5 Matches Returned (2.4ms) • Zero False Positives

Trusted by leaders across enterprise & academia

Vodafone Philips London Business School University of Cambridge Judge Business School UCL Cyprus International Institute of Management InnovateUK BridgeAI The Alan Turing Institute UK Export Academy Lloyd's Maritime Academy Electi Consulting Kalgera Office of Naval Intelligence Qualifications Wales British Land European Central Bank Square Trade Asfari Foundation Avasa AI Nile University Movemeback Greenhouse Intelligence Re_Skinned BankX
Who this is for

A specialist track, not another beginner library.

This is for people who already have a data, standards or AI problem that taxonomies and vector search cannot settle. Everyday AI courses live on the open-courses page. This page is the knowledge-graph line.

🏗️

Knowledge Engineers & Data Architects

Model complex domains, validate graph data at scale, and keep ontologies maintainable across enterprise lifecycles using RDF, OWL 2, SPARQL, and SHACL.

📐

Standards & Domain Leads

Join disparate industrial vocabularies across construction, insurance, space, and government without silent false friends, proving the bridge can be refuted.

🤖

AI & Governance Teams

Give AI agents and GraphRAG pipelines the explicit semantics, constraints, and reasoners required to prevent hallucinations and enforce auditability.

The core path

Two ways in. Same engineering standard.

Foundations is the recommended start if you want the W3C stack in practice. The complete course is the broader atlas: method, upper ontologies, domain standards and the 2026 AI overlay.

Ontology Foundations course cover
Recommended Start • 15 Lessons

Ontology Foundations (RDF, OWL and SHACL)

Progress from taxonomy to TBox and ABox modelling. Design a domain ontology, query it with SPARQL, validate it with SHACL, and connect it to GraphRAG and agent context.

  • RDF Data Model, Triples & Namespaces
  • OWL 2 Description Logics & Class Hierarchies
  • SPARQL Query Language & Graph Inspection
  • SHACL Data Shapes & Constraint Validation
  • GraphRAG Integration for Generative AI
Begin Foundations →
Complete Course in Knowledge Graphs cover
The Master Atlas • 45 Lessons • ~23 Hours

The complete course in taxonomies, knowledge graphs and AI

From what a category is to why a reasoner hangs. Covers RDF, OWL 2, SPARQL, SHACL, SKOS, upper ontologies (BFO, DOLCE, gist, SUMO, BORO, HQDM, IES), domain standards, GraphRAG and neurosymbolic method.

  • Full Upper Ontology Spectrum (BFO, DOLCE, gist, SUMO)
  • Neurosymbolic AI & Reasoning Engines
  • Formal Metamodelling & SKOS Vocabularies
  • Enterprise Portfolio Project & Validated Artefact
Begin the complete course →
Applied tracks

Same method, on a problem that already exists.

Each of these is open access, with a quiz on every lesson. Pick the domain you already work in rather than collecting the set.

Construction • 15 Lessons

IFC, Uniclass, COBie and BOT

Why construction data does not join up, how false friends break AI classification, and how to use an open published crosswalk on your own project.

Challenge: Siloed CAD/BIM schemas vs asset maintenance lifecycles.
Delivered: W3C BOT & IFC crosswalk with automated consistency proofs.
Open the construction course →
Curriculum Blueprint 100% Open Access

Construction Data Standards & AI

  • 🔹 Module 1: BIM Schemas (IFC4 & Uniclass 2015)
  • 🔹 Module 2: Topology Ontologies (W3C BOT & Brick)
  • 🔹 Module 3: Resolving Spatial Semantic Conflicts
  • 🔹 Module 4: Live Crosswalk Validation Suite
Format: Self-Paced Includes: Lesson Quizzes
The work behind the teaching

Published artefacts, not anonymous case studies.

These are independent demonstrations and open research, cited here because you can inspect them. They are not client outcome claims for a paid ontology programme.

51,355 triples

National Skills Ontology

Skills England occupational maps rebuilt as a graph that validates with zero SHACL violations, with all 1,269 standards placed.

Read the GovTech case →
NDTP · Apache-2.0

National Digital Twin Programme

An open-source AI ontology extension tool delivered for the Department for Business and Trade, plus published crosswalks between UK government ontology standards.

See the public research →
33,924 rows

European Insurance Register

An open ontology tested against every EU and EEA authorised insurer and reinsurer, then joined to GLEIF. The course teaches the same method used to find identifier defects in that study.

Clone the insurance artefacts →
For Organisations & Teams

If the problem is a live register or a live standard, start with a conversation.

The open courses teach the method. Advisory work applies it to one bounded dataset, crosswalk or agent workflow: meaning, constraints, evidence and human review.

Talk to us about ontology work → Explore AI Agent Governance

Looking for practical AI for everyday work, or the four-week AI Fluency Certification? Those live on open courses and AI Fluency. This track does not replace them.