PhD defence

From Text to Knowledge: Knowledge Graph Extraction and Ontology Learning with Language Models

• R.M. Bakker

Date

Tuesday 3 November 2026

Time

14:30 - 15:30 hour

Address

Academy Building
Rapenburg 73
2311 GJ Leiden

Supervisor(s)

• Prof.dr. S.A. Raaijmakers
• dr. M.H.T. de Boer

Summary

Human knowledge is vast. It is laid down in books, reports, legal documents, scientific papers, and, for the past decades, across millions of webpages. While this wealth of information is valuable, it is also overwhelming. No single person, or even a team of experts, can read and understand it all. To make sense of it, we increasingly rely on digital systems that organise information.

Knowledge graphs and ontologies help these systems understand how concepts are related. They are used in applications such as legal decision support and healthcare. However, creating and maintaining such knowledge models is still largely a manual process, making them expensive and difficult to keep up to date.

This thesis investigates how different types of language models and Natural Language Processing techniques can help automate the construction of knowledge graphs and ontologies from text. The research focuses on applied use cases and the practical challenges they present.

With the results from the different chapters, we show that language models are changing the way knowledge models are constructed. They can already extract relevant information from text and reduce the manual effort required to create structured knowledge. At the same time, abstract concepts and logical patterns remain challenging, meaning that human expertise remains essential for creating high-quality knowledge models. Rather than replacing experts, language models can serve as powerful assistants by suggesting relevant knowledge from collections of text. Ultimately, this can support professionals in making better use of the vast amounts of information available to them.

PhD dissertations

Approximately one week after the defence, PhD dissertations by Leiden PhD students are available digitally through the Leiden Repository, that offers free access to these PhD dissertations. Please note that in some cases a dissertation may be under embargo temporarily and access to its full-text version will only be granted later.

Press enquiries (journalists only)

[phone]
[e-mail]

General information

Beadle's Office
[e-mail]
[phone]