Ring-shaped Calorimetry Information for a Neural EGamma Identification with ATLAS Detector

This work concerns the identification process of electrons based only on calorimeter information. It is proposed the usage of ring-shaped description for a region of interest of the calorimeter which explores the shower shape propagation throughout the ATLAS calorimeters. This information is fed into a multivariate discriminator, currently an artificial neural network, responsible for hypothesis testing. The concept is evaluated for online selection (trigger), used for reducing storage rate into viable levels while preserving collision events containing desired signals. Preliminary results from Monte Carlo data point out that the background rejection can be reduced by as much as 50 % over the current method used in the High-Level Trigger, allowing for high-latency reconstruction algorithms such as tracking to run over at a later stage of the trigger.