Discovering and Characterising Exoplanets with the Next Generation Transit Survey

The Next Generation Transit Survey (NGTS) is a ground-based transiting exoplanet survey, whose aim is to discover Neptune and sub-Neptune sized planets orbiting bright, late-K and early M-type host stars (mv < 13). In fulfilment of the exoplanet and auxiliary science yield of the project, I characterise the mass, radius and orbital parameters of two interesting systems discovered: (1) An M-dwarf binary system with fully convective atmospheres; (2) A Jupiter sized planet in a 4.5 day orbit (Hot-Jupiter). I achieve this via Markov Chain Monte Carlo (MCMC) global modelling of photometric lightcurves and Radial Velocity (RV) curves. In addition, I show that the exoplanet candidate vetting process can be made more efficient by automatically vetting candidates with a Convolutional Neural Network (CNN) prior to human scrutiny.