About our computing science and mathematics research

Computing Science and Mathematics combines cutting-edge, fundamental research in predictive modelling and data science with their application in multiple domains.

We foster a dynamic interdisciplinary research environment in which computer scientists and mathematicians work together on solving challenging problems in food and health, the environment, business and social organisation.

Find out more about our research groups and PhD opportunities.

Abstract image of data flow

Research themes

Our research themes are:

  • Optimisation techniques (particularly meta-heuristic) where we apply applications to model parameter fitting, automated software engineering, scheduling problems.
  • Data-driven predictive modelling that melds traditional mathematical modelling with computational and data analytic approaches.
  • Vision and image processing where we devise new methods for extracting useful information from imaging data.
  • Food security is a major application area for data-driven predictive modelling, in combination with data science, remote sensing and image processing. This included work on strategies for controlling pest and disease spread.
  • Improving health and well-being where we use data science, image and sensor processing and predictive modelling for understanding health risks and providing solutions.
  • Cybersecurity for multimedia: to ensure legitimacy of electronic data, particularly images and videos.

Our research highlights

We use multimodal data analysis for monitoring aquatic invasive weeds in India. Predictive modelling and image analysis to predict the extent and impact of water hyacinth whilst community engagement methods assess the impact of weed infestation on local people.

Antimicrobial resistance is one of the top ten threats to global health, as identified by WHO. Key to minimising this risk is optimising the use of antibiotics. We are using predictive modelling and multi-objective optimisation techniques applied to scheduling of treatment to minimise the impact of antimicrobial resistance.

State-of-the-art signal processing techniques are being combined with exponential analysis with data driven approaches in the signal/image domain to improve sensing data analysis. This can address problems as diverse as antenna array design to malaria risk detection where the challenge is to identify small water bodies from remote sensing data. We are working with researchers across the world to advance exponential analysis and its applications.

We are exploring the use of deep machine learning and search techniques for developing non-linear partial differential equation models of tumour growth in lungs and their detection in CT scans for personalised, early lung cancer diagnosis.

We are collaborating with members from Biological and Environmental Sciences and Aquaculture in aquatic food security. We’re combining social, economic and ecological research to analyse the challenges associated with maintaining sustainability at the interface of people, their food supply and the environment.

We are also collaborating with members from Psychology (FNS) and Finance (Stirling Management School) in contextual learning and processing in humans and machines. This research aims to develop more effective intelligent decision-support systems powered by artificial intelligence and machine learning algorithms.

We are targeting our research in AI and machine learning at assistive technologies for dementia sufferers and the elderly.

Our applications are detecting fake news using fundamental techniques for multi-media cybersecurity.

Fundamental research is being carried out in signal processing to improve earth observation data analysis capabilities.

The development and analysis of new differential equation models is optimising ultrasound shock wave therapy to treat patients with kidney stones. It also helps predict the likelihood of potential heart failure.

Metaheuristic approaches were used to optimise and improve software used by Air France-KLM to test the robustness of their global flight schedules. This software is now faster and more accurate, due to this optimisation. Related optimisation approaches are being applied to the automation of software engineering with improved performances for businesses such as BT and Janus Rehabilitation Centre, Iceland.

Research by BIOMOD has used data-driven predictive modelling to identify the effectiveness of intervention strategies in pest management and food distribution. Specifically, modelling the spread of Louping III virus (LIV) determines the impact of host culling on the spread of ticks that carry the virus. These results are influencing policy on hare culling in Scotland.

Our partnerships

Our research is based on collaboration with a network of Scottish and international partners, universities, business and public bodies.

We work with an extensive Industrial Advisory Board of local businesses. We are active members of Scottish Informatics and Computer Science Alliance (SICSA) as well as working with Scottish DataLab and CENSIS Innovation Centres.

Our business partners include BT, KLM-Air France, Royal Botanic Gardens Edinburgh, DEFRA, Forestry Commission, Phonak AG, Fera Science Ltd, Bambu (Singapore), Scottish Rural College (SRUC), Unitech Ltd, UCare Foundation, Marine Science Support and others.