Skip to main content

Artificial Intelligence for Decolonising

Artificial Intelligence for Decolonising

Eric Atwell (School of Computing)

Project overview

The university is investing heavily in decolonising, through its key principles.

I will apply artificial intelligence and learning analytics to our teaching and learning data, to identify and measure colonial bias in our curriculum. This includes automated review of course reading lists in different disciplines to measure their colonial bias and diversity. It is widely assumed that decolonizing only applies to a few disciplines; I will seek to analyse teaching and learning data from a range of schools and faculties, to show whether “decolonising” applies beyond humanities and social sciences.

The research approach

I will employ the Cross-Industry Standard Process for Data Mining (CRISP-DM) methodology: this breaks down a project into 6 phases, which interact and can be repeated:

(1) Business Understanding, pinning down the terminology, objectives and requirements

(2) Data Understanding, collecting and annotating or labelling Reading Lists and other teaching and learning data

(3) Data Preparation for analysis with Machine Learning and Learning Analytics tools

(4) Modeling to learn features and indicators of colonial bias in our teaching

(5) Evaluation, eliciting feedback on usability and usefulness of the methods and results

(6) Deployment, publishing the tools and results for practical applications

The university is investing heavily in decolonising, this research will identify how decolonising applies across all subject and how we can find evidence in the teaching and learning data of colonial bias across a range of disciplines, this will help to justify the strategy.

Also, the project will contribute to Artificial Intelligence and Computing research. The School of Computing achieved Leeds University’s highest REF’2021 research GPA score (3.62/4). We will generate research publications and attract AI research students contributing to Leeds University’s next REF success. The learning data collected for the project will be a novel Gold Standard data-set for further Machine Learning and Text Analytics research, for example SemEval shared-task research initiatives.

If you would like to find out more about the project contact Eric Atwell (

Each fellowship has a project sponsor that helps the fellows achieve impact across the institution. The sponsor for this fellowship is Louise Banahene.

Project start date: 1 September 2022