We present the study of a fuzzy clustering algorithm for the Belle II electromagnetic calorimeter using Graph Neural Networks. We use a realistic detector simulation including simulated beam backgrounds and focus on the reconstruction of both isolated and overlapping photons. We find significant improvements of the energy resolution compared to the currently used reconstruction algorithm for both isolated and overlapping photons of more than 30% for photons with energies E < 0.5 GeV and high levels of beam backgrounds. Overall, the GNN reconstruction improves the resolution and reduces the tails of the reconstructed energy distribution and therefore is a promising option for the upcoming high luminosity running of Belle II.
@article{arxiv.2306.04179,
title = {Photon Reconstruction in the Belle II Calorimeter Using Graph Neural Networks},
author = {F. Wemmer and I. Haide and J. Eppelt and T. Ferber and A. Beaubien and P. Branchini and M. Campajola and C. Cecchi and P. Cheema and G. De Nardo and C. Hearty and A. Kuzmin and S. Longo and E. Manoni and F. Meier and M. Merola and K. Miyabayashi and S. Moneta and M. Remnev and J. M. Roney and J. -G. Shiu and B. Shwartz and Y. Unno and R. van Tonder and R. Volpe},
journal= {arXiv preprint arXiv:2306.04179},
year = {2024}
}