Tests of lepton universality using the RK* ratio

Using Run 1 data collected by LHCb the RK ratio was measured.
This measurement is particularly challenging as electrons undergo a lot of bremsstrahlung in the detector resulting in worse momentum and thus mass resolution for the B meson. If bremsstrahlung occurs before LHCb's magnet the radiated photon will not travel in the same direction as the electron so the photons will not hit the same calorimeter cluster. This required algorithms to be developed to find bremsstrahlung photons which should be added to a given electrons calorimeter cluster thus getting back some of the missing energy.
The aim of my research is to repeat and improve upon this measurement, in particular looking at training an MVA to reject partially reconstructed background which is a hinderance for the electron channels and looking into resampling the particle identification information obtained from our simulation with correlations intact. The goal is use what has so far been gathered of the Run 2 dataset to push down statistical uncertainty and potentially increase the discrepancy seen. This along with upcoming results from other analyses such as RK and Ro should pin down whether what has been seen is a genuine case of lepton universality violation or a statisical fluctuation.