WG211/M24Laemmel

Slides

PDF (on Google Drive)

Title

Type Inference in a Knowledge-Graph Setting

Abstract

What’s the framework of attribute, context-free and regular grammars to the programming language researcher, that’s the framework of description logics to the semantic web researcher. What’s type checking to the programming language researcher, that’s schema-based validation (think of SHACL and friends) to the semantic web researcher. What’s more or less functional code to the programming language researcher, that’s graph queries to the semantic web researcher.

Arguably, more than two worlds are at play -- especially when we talk about knowledge graphs and queries. That is, the database world is also at play, but let's focus on programming languages (PL) and semantic web (SW) here.

PL and SW are notably massively different:

• CWA is more Ok for the PL world; OWA is more Ok for the SW world, but both worlds aren’t shy to go back and forth.
• The PL world focuses on functions and procedures; the SW world focuses on queries, but queries aren't alien to PL.
• These two worlds also differ in terms of major pragmatics: (im)mutability, persistence, …

In this slot, we discuss “type inference in a semantic web / knowledge-graph setting” so that:

• we demonstrate inference of schemas for results of graph queries;
• we exemplify the challenges due to the open-world assumption for the data at hand;
• we pinpoint the role of description logics in formalizing type checking / inference in this setting.

Authors

Combined author team in recent years — in alphabetical order:

• Daniel Hernandez — University of Stuttgart
• Ralf Lämmel — University of Koblenz
• Philipp Seifer — University of Koblenz
• Steffen Staab — University of Stuttgart

Contributors regarding the broader theme:

• Martin Leinberger — Bosch Corporate Research
• Xinyi Pan — University of Stuttgart
• Tjitze Rienstra — Maastricht University
• Claudia Schon — Trier University of Applied Sciences

Further reading

• From Shapes to Shapes: Inferring SHACL Shapes for Results of SPARQL CONSTRUCT Queries
  • https://arxiv.org/abs/2402.08509
  • https://dl.acm.org/doi/10.1145/3589334.3645550
• From Shapes to Shapes: Inferring Shapes for Results of Construct Queries on Knowledge Graphs
  • To be submitted.

Also:

• eSPARQL: Representing and Reconciling Agnostic and Atheistic Beliefs in RDF-star Knowledge Graphs
  • https://arxiv.org/abs/2407.21483
  • https://iswc2024.semanticweb.org/event/3715c6fc-e2d7-47eb-8c01-5fe4ac589a52/websitePage:870dee1d-a340-45f2-af9c-bee1c707fbda
• ProGS: Property Graph Shapes Language
  • https://arxiv.org/abs/2107.05566
  • https://link.springer.com/chapter/10.1007/978-3-030-88361-4_23
• Deciding SHACL Shape Containment Through Description Logics Reasoning
  • https://arxiv.org/abs/2008.13603
  • https://link.springer.com/chapter/10.1007/978-3-030-62419-4_21
• Semantic Query Integration With Reason
  • https://arxiv.org/abs/1902.00545
  • https://programming-journal.org/2019/3/13/
• Type Checking Program Code Using SHACL
  • https://arxiv.org/abs/1907.00855
  • https://link.springer.com/chapter/10.1007/978-3-030-30793-6_23

See https://dblp.org/pid/l/RalfLammel.html for more.