$\nu_e$ signal selection and cross-checks performed using muon removed simulations and cosmic muon brem showers in NOvA

NOvA is a long-baseline neutrino oscillations experiment, designed to make precision measurements of $\theta_{23}$, $\delta_{CP}$, and the neutrino mass hierarchy using $\nu_\mu(\bar{\nu_\mu})\rightarrow\nu_e(\bar\nu_e)$ neutrino oscillations. It consists of two functionally equivalent detectors each located 14.6 mrad off the central axis of the Fermilab NuMI neutrino beam; the world's most intense neutrino beam. The Near Detector(ND) is located 1 km downstream from the beam source, and the Far Detector(FD) is located 810 km away in Ash River, MN. To validate the performance of our simulation of the convolutional neural network classifier used to identify $\nu_{e}$ events, we apply a data-driven technique called Muon Removal (MR). This talk will discuss two applications of MR. In a Muon-Removed Electron-added (MRE) study, we compare the performance of the classifier on ND $\nu_{\mu}$ CC candidates from both data and simulation in which the muon candidate is replaced by a simulated electron of the same energy. In Muon-Removed Bremsstrahlung (MRBrem), we remove the muon from data & simulated FD cosmic muon rays, resulting in a pure selection of electromagnetic showers. The resulting comparisons show agreement at the few-percent level.