Search for B → K$\nu \overline{\nu }$ decays with a machine learning method at the Belle II experiment

This thesis documents a search for the rare decay of a B meson into a K meson and a pair of neutrinos at the Belle II experiment, which is located along the SuperKEKB energy-asymmetric electron-positron collider. This decay has never been observed, its branching fraction is predicted with accuracy in the standard model of particle physics, and is a good probe of physics beyond the standard model. A novel method to search for this decay, the inclusive tagging, is developed on a data sample corresponding to an integrated luminosity of 189 fb${}^{-1}$ collected at the Y(4S) resonance, and a complementary sample of 18 fb${}^{-1}$ collected 60 MeV below the resonance. For this integrated luminosity, the expected upper limits on the branching fraction of B${}^{+}$→K${}^{+}$$\nu \overline{\nu }$ and B${}^0$→K${}_S^0$$\nu \overline{\nu }$ are determined from simulation to be 1.0×10${}^{-5}$ and 1.8×10${}^{-5}$ at the 90% confidence level, respectively. When the method is applied to data samples of 63 fb${}^{-1}$ collected at the Υ(4S) resonance and 9 fb${}^{-1}$ collected 60 MeV below the resonance, no significant signal is observed, and an upper limit on the branching fraction of B${}^{+}$→K${}^{+}$$\nu \overline{\nu }$ is determined to be 4.1×10${}^{-5}$ at the 90% confidence level.