We have another exciting Geographic Data Service PhD opportunity based at the University of Liverpool.
Project Overview
Visualizing Place: Using Multimodal AI to Describe and Represent Geodemographic Classifications
This PhD project will explore the application of large language models (LLMs) and vision language models (VLMs) to enhance the description, interpretation and visualization of geodemographic classifications. The research will investigate how generative AI can be leveraged to provide meaningful textual narratives and visual representations that characterise geographic areas, moving beyond traditional numeric profiling approaches. The project bridges geodemographics, natural language processing, computer vision, and multimodal AI, with a particular focus on how VLMs and text-to-image models can generate representative visual archetypes that make geodemographic insights more accessible and engaging.
Research Questions
- How can LLMs be used to generate accurate, contextually appropriate narrative descriptions of geodemographic clusters?
- How can multimodal AI (combining LLMs and VLMs) improve the interpretability of complex demographic and geographic data?
- What are the limitations, biases, and ethical considerations when deploying generative AI for geodemographic characterisation?
- How can these approaches be operationalised for policy and planning applications?
Potential Research Avenues
LLM-based Narrative Generation
- Fine-tuning LLMs (or using prompt engineering) to generate coherent descriptions of geodemographic archetypes based on underlying data
- Evaluating output quality, factual accuracy, reproducibility and utility for stakeholder communication
- Comparing outputs across different LLM architectures and parameter sets
Vision Language Model Integration
- Using VLMs to analyse satellite imagery, aerial photographs, and street-level data
- Extracting semantic features (urban density, vegetation, built environment characteristics) and linking to geodemographic profiles
- Generating typical descriptive images for each geodemographic class using text-to-image models, creating visual archetypes that complement textual descriptions
Visualization and Generative Imagery
- Using text-to-image models to generate representative visual archetypes for each geodemographic classification
- Developing prompt engineering strategies to ensure generated images accurately reflect cluster characteristics (housing types, density, greenspace, socioeconomic indicators)
- Evaluating stakeholder responses to AI-generated visual representations versus traditional statistical charts and maps
- Investigating how generated imagery might reinforce or challenge stereotypes and biases in geodemographic representation
- Creating interactive visualization tools that combine traditional geodemographic maps with AI-generated imagery and narrative descriptions
Methodological Considerations
- Data integration: combining traditional geodemographic variables with AI-derived interpretations from unstructured data
- Bias assessment: identifying and mitigating representational, algorithmic, and output bias in generative models
- Explainability and transparency: ensuring outputs are interpretable and defensible for policy use
- Reproducibility: documenting prompts, model versions, and hyperparameters
Policy and Applications
- Collaboration with policymakers to prototype tools for communicating geodemographic insights
- Testing approaches with urban planning, public health, and transport agencies
- Developing best practices and guidelines for responsible use of AI in geodemographic research
Candidate Profile
Candidates will have, or be due to obtain, a Master’s Degree or equivalent in a relevant subject. Exceptional candidates with a First Class Bachelor’s Degree in an appropriate field or significant relevant experience will also be considered. Ideal candidates will possess:
- Strong foundation in geography, data science, or related quantitative discipline
- Proficiency in Python and experience with GIS tools (GeoPandas, QGIS, Google Earth Engine)
- Familiarity with LLM and generative AI APIs (OpenAI, Anthropic, Hugging Face, Stable Diffusion, DALL·E)
- Understanding of machine learning, natural language processing, and/or computer vision
- Interest in research ethics, AI bias, and responsible AI development
- Excellent written and verbal communication skills
- Experience with urban analytics, geodemographics, data visualization, or policy-engaged research is a plus
Supervision and Support
This project is based within the Geographic Data Service at the University of Liverpool, and will be supervised by Professor Alex Singleton.
Funding and Entry Requirements
Applicants should hold (or expect to hold) a first-class or upper second-class honours degree in geography, computer science, data science, environmental science, or a related discipline. Applicants from other quantitative backgrounds with strong transferable skills are encouraged to apply.
IMPORTANT – The studentship is fully funded, including academic fees up to the UK rate (international applicants would be required to top-up the difference) – https://www.liverpool.ac.uk/study/fees-and-funding/tuition-fees/postgraduate-research/), a stipend (2026-27 rate is £21,805 per year), a £5k research training support budget, and is available to start from October 2026.
Funding and Entry Requirements
Stage 1: Deadline 21st March 2026.
Please email a one‑page statement outlining your interest in the project and your CV to Professor Alex Singleton (alex.singleton@liverpool.ac.uk) no later than 21st March 2026
Use the subject line “PhD Application – Visualizing Place”. If you are a suitable candidate, you will be asked to apply formally to the University of Liverpool, and details will be provided.
Stage 2
Candidates selected for Stage 2 will be invited to an online interview and presentation to discuss their research ideas, alignment and general interest in the PhD project.
Related Publications
The following are related publications to this PhD:
Alex Singleton; Seth E. Spielman (2026). Computers, Environment and Urban Systems, 125, 102396. DOI: 10.1016/j.compenvurbsys.2025.102396
Alex Singleton; Seth E. Spielman (2024). EPJ Data Science, 13(1). DOI: 10.1140/epjds/s13688-024-00466-1
Alex Singleton; Dani Arribas-Bel; John Murray; Martin Fleischmann (2022). Computers, Environment and Urban Systems, 95, 101802. DOI: 10.1016/j.compenvurbsys.2022.101802
