Tutorial: Using SHACL and OWL in
Combination for Practical Knowledge Graph
SCOPE Tutorial: OWL & SHACL in Practice
Knowledge graphs power enterprise data integration, semantic backbones and AI applications, but combining OWL’s open-world reasoning with SHACL’s closed-world validation is still a real challenge in day-to-day projects. This half-day, hands-on tutorial shows you how to use both standards together for intuitive ontology and instance data editing, consistent data quality at every step, and GenAI-assisted, conversational ontology editing.
Who it’s for: Intermediate knowledge graph practitioners with prior OWL and SHACL knowledge who want practical guidance for real-world projects.
Key takeaways
- Combine OWL and SHACL correctly, staying true to W3C standards.
- Keep data consistent across the whole data engineering process.
- Speed up ontology work with GenAI, including conversational editing.
- Model a real use case hands-on, from data model to instance data.
- Prepare ontologies for downstream use, such as GraphRAG.
Agenda
- 09:00 – 09:45 | Part 1: OWL and SHACL Challenges in Practice
- Intro to how OWL and SHACL are used in real-world applications, and how they can complement each other.
- 09:55 – 10:45 | Part 2: Setting Up the Tools and Bootstrapping an Ontology
- Get the tools running and kick-start an ontology with GenAI support.
What you need to bring:
Install Docker on your machine. This is mandatory to use any tool of our toolset.
Have an LLM account ready. You can use any LLM, but Claude is recommended.
Install a tool/command to let your LLM client call MCP servers. e.g. npx (to be used with mcp-remote)
📄 Bring your own documents to bootstrap an ontology via our tooling pipeline! - 11:00 – 12:00 | Part 3: Hands-on Data Modelling
- Model your use case’s ontology, starting either from the bootstrapped ontology or let you guide by our GenAI workflows.
- 12:10 – 12:50 | Part 4: Discussion & Feedback
- Share your experience, ask questions, and compare approaches with peers and experts.
