Developing a dashboard for social and environmental risks using data mining and multi-criteria decision analysis

Background

The social and environmental implications of climate change are not equally distributed which means that there will be some geographic areas at higher risk than others. It is therefore important to collect data from various disparate sources to assess different types of risks, and then, using multi-criteria decision-making techniques to combine and present this information to policy makers in a comprehensible manner. The aim of this work is to develop a dashboard to visualise and identify high-risk areas on geospatial map, and eventually to help policy makers in making evidence-based decisions.

PhD opportunity

This project will have a number of questions to investigate. The first challenge would be to understand various dimensions and types of risks attached to climate change and identifying data sources that can help quantify these risks. This might include techniques from data science (like web scraping, text mining, data wrangling, information fusion etc.). Once the data are cleaned and transformed into quantifiable risk scores, the next challenge would be to merge different types of risk into a single overall risk score. This will involve the use of multi-criteria decision making and preference elicitation methods. This project will offer students to learn several new tools and techniques from different academic disciplines, and yet, applying them for a single problem at hand.

Other information

Applicant profile: Strong background in statistics and data science skills (e.g. data collection, data cleaning, joining data from different sources, handling missing data, dashboard development).

  • https://www.metoffice.gov.uk/about-us/news-and-media/media-centre/weather-and-climate-news/2023/new-climate-data-portal-makes-data-more-accessible
  • https://doi.org/10.1109/ICDMW.2008.30
  • https://doi.org/10.1016/j.eiar.2024.107607
  • https://doi.org/10.3390/w13101358
  • https://doi.org/10.3390/w12092379