PhD position in modelling and representation of geographic questions

Join the ERC-funded GeoTrAnsQData project and explore hybrid AI approaches to better understand, structure and formalise geo-analytical questions. This helps shape the future of geospatial reasoning and map-based knowledge discovery.

Your job

Geographic questions like 'What is the potential for reducing urban heat in Amsterdam by installing green roofs on existing buildings?' are important in fields such as urban planning, sustainability, and public health. Answering such a question requires the transformation of maps combining different suitable geodata sources, including heat sources and building layouts to generate an answer map. This problem is called indirect question answering, and it is not straightforward with current GeoQA tools. In such scenarios, maps must be created or transformed from other maps.

The ERC funded project GeoTrAnsQData project addresses this by developing a GeoQA method that converts questions into executable geo-analytical workflows, turning geodata into new answer maps accordingly. We use knowledge graphs to model these transformations and apply AI methods to scale them up across large map repositories, enabling users to explore many ways maps can be reused to answer different kinds of questions. This PhD position focuses on understanding, interpreting, parsing and formalising geo-analytical questions. You will explore hybrid (symbolic and sub-symbolic) AI approaches to help users formulate and translate natural language questions into structured representations that can be linked to geospatial data sources and workflows.

You will contribute to the design of the linguistic and conceptual interface between natural language questions and formal workflow models over a geodata repository. In this project, you will:

• build and annotate a corpus of geo-analytical questions and their associated purposes, data needs, and analytical steps in geo-analytical standard scenarios;
• develop a model of geo-analytical purposes (transformation requests) and a corresponding question grammar;
• perform a user study on geo-analytical question formulation to express such purposes;
• contribute to the formalisation of spatial question types using a purpose-driven taxonomy;
• develop a hybrid question parsing pipeline using NLP and formal semantic representations; investigate Large Language Models (LLMs) as well as symbolic AI for question parsing;       
• evaluate models based on a gold standard of geo-analytic purposes and questions;
• collaborate with a technical assistant, another PhD candidate (on geodata source modeling), and a postdoc (on the GeoQA reasoning engine).

This position is ideal for someone interested in natural language processing, geographic information and knowledge representation. It is part of the ERC-funded project GeoTrAnsQData, which develops the foundations of a transformative GeoQA methodology through an integrated research program across geoinformatics, AI, and geography. The project is based at the Department of Human Geography and Spatial Planning, Utrecht University, and contributes to cutting-edge research on spatial reasoning, semantic technologies, and interdisciplinary AI for geosciences and geography.

Your qualities

• A (near-)completed MSc degree in Artificial Intelligence, Computational Linguistics, NLP, GeoInformatics, Geography, Geoscience, or a related field;
• experience with natural language processing (NLP), language modeling, grammatical models, and question-answering (QA);
• a strong interest in maps and geographic information as well as spatial questions;
• interest in conceptual modelling, knowledge representation and knowledge graphs;
• programming skills in Python and knowledge of relevant NLP libraries (e.g. spaCy, Hugging Face Transformers); 
• good English communication skills and an interest in interdisciplinary collaboration.

Our offer

• A position for one year, with an extension to a total of four years upon a successful assessment in the first year, and with the specific intent that it results in a doctorate within this period;
• a working week of 38 hours and a gross monthly salary between €3,059 and €3,881 in the case of full-time employment (salary scale P under the Collective Labour Agreement for Dutch Universities (CAO NU));
• 8% holiday pay and 8.3% year-end bonus;
• a pension scheme, partially paid parental leave and flexible terms of employment based on the CAO NU.

In addition to the terms of employment laid down in the CAO NU, Utrecht University also offers a range of its own schemes for employees. This includes arrangements for professional development, various types of leave, and options for sports and cultural activities. You can also tailor your employment conditions through our Terms of Employment Options Model. In this way, we encourage you to keep investing in your personal and professional development. For more information, please visit Working at Utrecht University.

About us

A better future for everyone. This ambition motivates our scientists in executing their leading research and inspiring teaching. At Utrecht University, the various disciplines collaborate intensively towards major strategic themes. Our focus is on Dynamics of Youth, Institutions for Open Societies, Life Sciences and Pathways to Sustainability. Sharing science, shaping tomorrow.

Utrecht University’s Faculty of Geosciences studies the Earth: from the Earth’s core to its surface, including man’s spatial and material utilisation of the Earth – always with a focus on sustainability and innovation. With 3,400 students (BSc and MSc) and 720 staff, the faculty is a strong and challenging organisation. The Faculty of Geosciences is organised in four Departments: Earth Sciences, Human Geography & Spatial Planning, Physical Geography, and Sustainable Development.

More information

For more information about the position, please contact Dr Simon Scheider at s.scheider@uu.nl.

Candidates for this vacancy will be recruited by Utrecht University.

Apply now

As Utrecht University, we want to be a home for everyone. We value staff with diverse backgrounds, perspectives and identities, including cultural, religious or ethnic background, gender, sexual orientation, disability or age. We strive to create a safe and inclusive environment in which everyone can flourish and contribute.

Knowledge security screening can be part of the selection procedures of academic staff. We do this, among other things, to prevent the unwanted transfer of sensitive knowledge and technology.

To apply, please send your curriculum vitae, including a letter of motivation, via the ‘apply now’ button.

The first round of interviews will take place in February or March 2026. The aimed starting date is 1 May 2026.

The application deadline is 31 January 2026.