Application of a Boosted Decision Tree to signals andbackgrounds of the search for Dark Mesons at theATLAS experiment

Abstract: An ongoing analysis at the ATLAS experiment is searching for a Dark Matter candidate in the shape of Dark Mesons. Simulations of the signal and background data have been produced with the Monte Carlo method. This project tests the possibilities of using Boosted Decision Trees for the distinction between the simulated signal and background data. A model has been trained on a set of various kinematic variables and an attempt was made to rank the quality of these variables for the analysis. The BDT was succesfull in distinguishing signal and background.