School of Computer Science and Statistics
This project would suit an undergraduate FYP, particularly if taking the 4th year Human Factors module. The aim of the project would be to replicate classic human performance experiments using computer games as a testbed, starting with simpler phenomena such as Fitts’ Law (representing the time taken to hit a screen target based on distance … Read more
Several qualitative research methodologies involve analysis of data such as interview transcripts and open questionnaire responses. For example, several forms of analysis involve coding of the transcripts and then identifying common patterns and themes within the codes. There are potential benefits as well as risks associated with the use of AI tools to support such … Read more
With tools such as Transkribus (https://www.transkribus.org/) and eScriptorium (https://www.sofer.info/), AI-based image processing techniques are unblocking what has been a hugely expensive and time-intensive process of turning historical documents into machine readable transcriptions, in a manner that enables a scaling up of the process.The next problem that then needs to be tackled is how to support … Read more
Using timelines to visualise critical data is particularly useful for domain experts to examine and analyse a wide range of data. For example for Historians a timeline visualisation such as seen in an early attempt of this (https://timelines.virtualtreasury.ie/timelines/) for the Virtual Record Treasury of Ireland team is really helpful. This user interface was developed a … Read more
This project will be in the area of Explainable AI (XAI), and specifically building of explanations to support learning and user interfaces for presenting explanations to the end user (learner). The focus will be on the explanations rather than machine learning technology, although the student can use the project as an opportunity to explore these … Read more
This project focuses on the rich field of algorithmic fairness where the goal is to ensure that predictions are not biased against subgroups of the population whilst maximising predictive performance. One challenge is when we focus on multiple protected attributes.
Smartwatches are increasingly used to track a variety of health-related data, but the focus is mainly on automated tracking. Recent years have seen an expansion in the interaction capabilities of smartwatches (e.g. using voice or gesture). This project will investigate different modalities for entering and annotating self-tracking information via smartwatch. The project will suit better … Read more