Master theses

Knowledge Graphs have become the interoperable backbone for exchanging structured information across systems, organisations, and domains — powering everything from Google's Search Knowledge Graph to the EU's data spaces initiatives. But most data in the world does not live in a KG natively: it lives in relational databases, CSV exports, REST APIs, and XML feeds. Bridging that gap requires a mapping layer, and that is exactly what RML — the RDF Mapping Language — provides.

RML is currently being standardized through the W3C Knowledge Graph Construction Community Group, a process that brings together researchers and practitioners from across Europe and beyond. Standardization is a critical step toward broad adoption, but it comes with a hard engineering challenge: a standard is only as credible and useful as its reference implementation. Without a solid, spec-compliant engine that the community can test against, ambiguities in the specification go undetected, interoperability suffers, and adoption stalls. The specification itself is also a moving target — it evolves rapidly as the community resolves open issues and adds new features, meaning any implementation must be designed to keep pace.

In this thesis, you will take a central role in the development of MappingWeaver, IDLab's reference Java implementation for the emerging RML standard. MappingWeaver is purpose-built to track the W3C specification closely — but there is significant ground left to cover: spec coverage gaps, engineering challenges around correctness and performance, and the meta-problem of how to structure a codebase so that it can evolve in lockstep with a living standard.

You will analyse the current state of the implementation against the latest spec drafts, design and implement missing features, and — crucially — develop a methodology and tooling to keep the implementation aligned as the specification continues to evolve. This includes working with the official RML test suite to drive compliance, contributing feedback from implementation experience back into the standardization process, and potentially influencing how the spec itself is written.

This thesis is a rare opportunity to do engineering work that directly shapes an international W3C standard. You will collaborate with IDLab researchers who are active participants in the KG-Construct CG, giving you a front-row seat in open standards development and a concrete, lasting artifact — a reference implementation used by the global KG community.

The scope is intentionally bounded to high-impact specification gaps and conformance evidence, rather than complete feature parity.