School of Computer Science and Statistics
This project will aim to generate synthetic data of children performing a set of motor skills, such us hoping, jumping forward, skipping etc. Real data sets of children performing the skills will be provided. The end objective is to generate ML models of the children performing the various motor skills tasks.
This project involves creating an app to anonymise (blur) large amounts of video footage of people wearing face masks. The obfuscation technique developed should have a minimal impact on the accuracy of vision recognition models. So, the anonymised data should keep markers to core reference points such as ears, eyes, nose etc. so that the … Read more
This project will look at how to assess LinkedIn profiles to identify whether the owner of the profile exhibits a set of employability skills such as reliability, common sense, commitment, motivation, enthusiasm, ability to deal with pressure, etc. The project will require an analysis of the LinkedIn features which could potentially allow users to exhibit … Read more
I am interested in any project that applies Artificial Intelligence and Machine Learning algorithms to real-time data gathered from sensors (or any Internet of Things devices) to address sustainability challenges in smart cities. The perspective of any stakeholder within a city is interesing to explore (citizens, commuters, bus providers, private transport providers (bike, taxi, car), … Read more
Clustering methods are usually applied to a single data set of interest. However, several related data sets can sometimes be available. In this case it is often of interest to determine whether the features of a single data set identified by a cluster analysis are common across all data sets. Consensus clustering has been proposed … Read more
Supervised with the assistance of Gary White. A digital twin is a highly accurate digital representation of a physical process, person, place, system or device. Digital twins were originally designed to improve manufacturing processes using simulations that have highly accurate models of individual components. However, with increasingly scalable and accurate BIM model creation capabilities, it … Read more
Problem/Background Effective citation parsing is crucial for academic search engines, patent databases and many other applications in academia, law, and intellectual property protection. It helps to identify related documents and calculate the impact of researchers and journals (e.g. h-index). “Citation parsing” refers to identifying and extracting a reference like “[4]” in the full text, and … Read more