Data Bytes 2: Space & Mobility

The 2nd instalment of our data science mini-series showcasing current research projects carried out by the Data Scientist Development Programme at Leeds Institute for Data Analytics (LIDA).

This week’s features 10-minute lightning talks from three data scientists on using data analytics to AI & Automation.

Each talk will last 10mins + 10mins Q&A.

AGENDA

Talk 1 - Evaluating the impact of cultural events using machine learning and digital footfall (Marion Carneiro)

Talk Summary - This project brings together the Centre for Cultural Value, The Audience Agency, and Bradford 2025 to support the development of a National Cultural Data Observatory (NCDO). Using Bradford City of Culture 2025 as a case study, it seeks to understand the impact of cultural activities on neighbourhood vitality. Mobile app-derived footfall data and machine learning models were used to estimate baseline footfall, identify event-related increases and explore post-event patterns. These insights will strengthen cultural impact assessment, support evidence-based decision-making for event organizers, funders and policy makers, and will be integrated into the NCDO.

Speaker Biography - Marion is a data scientist with a background in environmental and sustainability analytics, applying machine learning and geospatial methods to address real-world policy and climate challenges. I'm motivated by projects that support informed decision-making for the public good and a sustainable future through innovative data solutions.

Talk 2 - Wheely Easy: Creating a Wheelability Indicator for Bradford (Chenrui Xiao)

Project Summary - Our research develops a methodological framework for constructing a wheelability network that captures wheelchair-specific mobility constraints. Using mobile LiDAR data and an initial classification model, we extract road geometry and apply a skeletonisation-based geometric workflow to detect dropped kerbs. Sidewalks are extracted and reduced to centrelines, from which slope and width metrics are used to compute a Wheelability index. Integrating these indices with dropped-kerb connectivity yields a comprehensive network representing real wheelchair navigability. By highlighting dropped kerbs as essential mobility elements, this work advances wheelability assessment beyond existing pedestrian-focused approaches.

Biography - Chenrui Xiao is a Data Scientist at the Leeds Institute for Data Analytics, working at the intersection of data science and urban spatial research. He is interested in how computational methods can reveal urban dynamics and mobility patterns, with current work focusing on LiDAR-based modelling of wheelchair mobility.

Talk 3 - Developing a Geospatial Foundation Model for Mobility (Damilola Ogungbemi)

Talk Summary - This project is an exploratory step towards a geospatial foundation model for mobility, examining whether representations derived from mobility structure can capture urban socio-demographic patterns. Using UK Census journey-to-work data for Leeds and London, we model commuting flows at the LSOA-level, comparing a classical gravity model against a self-supervised graph representations learned model via Deep Graph Infomax. These models are then evaluated via the downstream tasks of inferring socio-demographic variables such as population and Index of Multiple Deprivation. We also consider whether mobility trajectories could provide a more robust basis for future geospatial foundation models.

Speaker biography - With a background in healthcare and an MSc in Biotechnology, Damilola Ogungbemi has published a research paper in cell biology, led interdisciplinary projects, and uncovered insights from complex datasets. Passionate about using data responsibly to create meaningful, positive impact.