I am a second year PhD student in the Astronomy and Astrophysics group at the University of Warwick, working under the supervision of Dr. David Armstrong. I completed my MPhys degree at the University of Warwick in 2019 and prior to commencing my PhD in 2020, I spent a year working as a Research Assistant in the Astronomy and Astrophysics group.
My research is currently focused on the development of an automated vetting and validation tool for exoplanet candidates from the TESS mission. At its core, the tool employs Machine Learning (ML) algorithms and Bayesian analysis to separate planetary candidates from false positive (FP) detections and validate those with high (>99%) posterior probabilites of being true planetary complanions. This is achieved by training the ML algorithms on a synthetic training set, where simulated transits occuring from planets and other astrophysical FP scenarios have been injected in TESS lightcurves. The alogrithms are then used to classify each candidate accordingly. The ML classification score is combined with computed prior probabilities to derive the posterior probability of the candidate being a true planet.
My past work as a Research Assistant involved the devlopment of an automated tool to map flooding events detected in radar observations from the Sentinel satellites of ESA's Copernicus Programme. This included the creation of an algorithm to detect flooded pixels in optical satellite images, which were used as visual ground truth, and the employment of ML algorithms for the classification of flooded pixels in the radar images. The tool was developed as a pipeline framework, taking satellite radar images before and during the flood as input, processing them to create and extract the necessary features, applying the ML classification and then writing the resulting scores as additional pixel-level data in the original images.