Machine learning for establishing ecological networks in agricultural landscapes

We have an exciting PhD position at the School of Computer Science and Electronic Engineering, University of Surrey. This multi-disciplinary PhD research project focuses on improving biodiversity risk assessment in agricultural landscapes by leveraging advanced AI and machine learning. Given the global urgency to halt biodiversity loss and promote sustainable food production (e.g., EU Biodiversity Strategy 2030), this project addresses a critical gap in understanding ecological interactions and their implications for biodiversity protection.

Currently, biodiversity assessments face challenges due to the lack of clear protection goals and effective methodologies. A key debate in biodiversity science is whether diversity should be assessed through species counts (taxonomic) or functional traits. This project aims to resolve this debate by evaluating the relative merits of these two assessment methods using advanced AI and machine learning.

Using data from our partner's BioAWARE project, which includes comprehensive species abundance and DNA metabarcoding data on carabid beetles and weeds, the research will build trophic networks and analyse ecological interactions. The novel approach lies in applying advanced AI for network inference and analysis, particularly Meta-Interpretive Learning (MIL), a logic-based machine learning technique that offers explainable and sustainable solutions by learning explicit ecological interaction rules more efficiently and with less data compared to traditional statistical machine learning methods. 

The project’s objectives include constructing species and functional networks, validating them against metabarcoding data, and developing new ecological rules using MIL. Deliverables include high impact publications and publicly available AI/computational tools. This research promises to advance ecological network learning/analysis, providing a powerful tool for biodiversity management and contributing significantly to sustainable agriculture.

You will be a core member of the Human-Machine Learning and Reasoning Lab and part of the Surrey Institute for People-Centred Artificial Intelligence and Institute for Sustainability.

Open to any UK or international candidates. Up to 30% of our UKRI-funded studentships can be awarded to candidates paying international rate fees.

You will need to meet the minimum entry requirements for our PhD programme.