- BSc (Hons) Experimental Physics 2007 University College Dublin
- PhD Solar Physics 2012 Trinity College Dublin
- Software Engineer Skytek 2012-2016
- Research Fellow DIAS/TCD 2016 - Present
- Lab Supervisor, 2019 - 2nd Year Physics Computational Labs
- Lab Supervisor, 2018 - 1st Year Physics Experimental Labs
- Lab Assistant, 2008-2010 - 1st Physics Experimental Labs
Finucane K. (2019) - Final Year Project - Solar Flare Studies Using Machine Learning
Current solar flare detection and classification relies on simple rule based systems. These systems can not deal with concurrent flares and struggle to identify very small flares. This project aimed to demonstrate Long Short Term Memory (LSTM) neural networks are an ideal candidates for flare detection and classification. An LSTM biased approach should also be able to over come the concurrent flare limitation and also improve detection sensitivity thresholds.
Power, D. (2018) - Final Year Project - Classifying Active Regions using Convolutional Neural Networks Sunspot classification is currently performed by human operators limiting the number of classification and introducing bias. The aim of this project was to demonstrate this task could be automated using machine learning, convolutional neural networks in particular. If sufficient accuracy could be achieved
Doherty, J. (2018) - Final Year Project - Statistical Analysis of Coronal Mass Ejection Velocity Oscillations This project aimed to study the statistical properties of coronal mass ejection (CME) velocity Oscillations. Such oscillations if present could be used to estimate the magnetic field strength of CMEs one of the most crucial properties in determining the geo-effectiveness of CMEs.
Speer, E. (2017) - MSc Thesis - Analysing the Presence of Coronal Mass Ejection Velocity Oscillations A number of recent studies have shown evidence of coronal mass ejection (CME) velocity oscillations. These oscillation are of fundamental scientific interest and but also in an operational sense when applied to space weather. This project aimed to extend the previous studies by increasing the number of events studied and applying more advanced fitting and statical analysis techniques.