Rejection of multi-jet background in a hadron collider environment through a SVM classifier

We optimized a multivariate discriminant software package, based on the Support Vector Machine (SVM) algorithm, to reduce the multi-jet background events in the channel pp̅ → ev + jj̅. This channel is important for many physics searches but the multi-jet background can be large and it is difficult to model. We developed a package which allows training set selection, maximization of efficiency and consistency checks. In this paper we will show how the multivariate approach we presented proved to be more efficient compared to the state of art approaches, both in terms of classification accuracy and background contamination.