Professor Alison Heppenstall

Profile

My PhD was a mixture of spatial econometrics and artificial intelligence, specifically building agent-based models to replicate dynamics within a retail market (petrol prices).   Subsequent postdoctoral work focused on the building of Machine Learning approaches such as neural networks and evolutionary algorithms for both flood and water quality prediction.  Whilst still working heavily with individual-based approaches (behavioural models in crime and retail), my research interests span a wide range of ML and AI approaches including probabilistic programming, graph theory, deep learning, reinforcement learning, emulators, particle filters, neural networks etc. These are used to understand and simulate complex spatial systems, in particular, cities.

I hold an ESRC-Alan Turing Institute Fellowship.

Responsibilities

• ESRC-Alan Turing Fellow
• Director of Innovation (CDRC)

Research interests

My research is centred around developing and applying methods from other disciplines to solving complex spatial problems particularly in the area of Urban Analytics.  I am interested in artificial intelligence, machine learning, data analytics and visualisation.  A current list of funded projects is below:

Bringing the Social City to the Smart City (ESRC-Turing Fellowship): This three year fellowship looks at (i) methodologies for uncovering hidden patterns and processes in spatio-temporal systems; (ii)  casual relationships in populations and (iii) explores uncertainty in individual-based modelling for city simulation.

Systems Science in Public Health and Health Economics Research (SIPHER) : SIPHER vision is a shift from health policy to health public policy.  Along with Dr Nik Lomax, I am responsible for the data management and micro-modelling work streams of this 5 year UKPRP consortium

Behavioural, ecological and socio-economic tools for modelling agricultural policy (BESTMAP - H2020):  My role in this project is to devise ways to scale up ABMs from local to national levels.

Consumer Data Research Centre (ESRC):The CDRC seeks to develop new approaches to social science research which are needed to exploit new sources of consumer data. I hold the post of Director of Innovation.

Below are Turing projects that I am involved with - more information can be found via my Turing page.

Understanding and Quantifying Uncertainty in Agent-Based Models for Smart City Forecasts: (Turing) Developing methods that can be used to better understand uncertainty in individual-level models of cities

Capturing relationships between individuals: Integrating Causal Inference and Agent-based modelling: (Turing). This project will connect ongoing work in casual inference modelling to agent-based simulations to robustly capture and simulate causal relationships between individuals.

Forecasting the future of policing (Turing): This project is in conjunction with UCL and The Met to explore the potential of ABM as a tool for forecasting demands in policing.  The PI is Dr Dan Birks (University of Leeds).

Quantifying Utility and Preserving Privacy in Synthetic Data (QUIPP):  This is a joint project with the Turing that is aims to generate synthetic versions of sensitive data sets that contain all the relationships and preserve individual privacy.

Real-Time Advanced Data assimilation for Digital Simulation of Numerical Twins on HPC (RADDISH): This project will perform the essential computational groundwork to allow researchers to apply DA methods to coupled human-environmental systems. The overall PI is Prof Serge Gullias (UCL).

Data Assimilation for Agent-based models (ERC): This project is devising new approaches to calibrate and validate ABMs in real-time, thereby improving the accuracy of short-term forecasts of social systems.  Model code and updates can be found on the following Github pages. The PI is Dr Nick Malleson.

• Bringing the Social City to the Smart City
• Centre for Computational Geography
• Geospatial restructuring of industrial trade (GRIT): integration of secondary data to model geospatial economic responses to fuel price
• Modelling Individual Consumer Behaviour
• System-science Informed Public Health and Economic Research for Non-communicable Disease Prevention (the SIPHER Consortium)

Qualifications

• PhD AI
• MSc GIS
• BA (Hons) Archaeology

Professional memberships

• Fellow of the Royal Geographical Society

Student education

I teach Geocomputation and GIS at both undergraduate and postgraduate levels.

Research groups and institutes

• Institute for Spatial Data Science

Projects

<li><a href="//phd.leeds.ac.uk/project/2063-dynamic-microsimulation-for-health-">Dynamic microsimulation for health </a></li>