Abstract:
The fundamental composition of and interactions of matter in the universe is described by a collection of quantum field theories known as the Standard Model (SM). Though the SM has been thoroughly tested at colliders such as Tevatron and the Large Hadron Collider (LHC), there remains strong motivation for physics beyond the SM or BSM. The LHC searches for BSM rely heavily on reconstructing and identifying clean and isolated electron trajectories. However, a class of BSM model predicts the very close or merged electron signatures in the detectors. A merged electron means that there is a huge overlap of clusters among the two electrons. One such model is the Right Handed Neutrino's (RHN). An SM counterpart of the above is a boosted photon or $Z$ boson giving close-by or merged electrons. All the current reconstruction algorithm fails to reconstruct the two individual electrons. In this study reconstruction of these merged electrons is studied. Multivariate analysis (MVA) techniques like a neural network (NN) classifier have been used to tag merged electrons. The NN classifier showed good performance in tagging these objects and separating them from genuine clean and isolated electrons.
