TMVA: The Toolkit for Multivariate Data Analysis eith ROOT

Multivariate classi cation methods based on machine learning techniques have become a fundamental ingredient to most physics analyses. The classi cation techniques themselves have also signi cantly evolved in recent years. Statisticians have found new ways to tune and to combine classi ers to further gain in performance. Integrated into the analysis framework ROOT, TMVA is a toolkit offering a large variety of multivariate classi cation algorithms. TMVA manages the simultaneous training, testing and performance evaluation of all the classi ers with a user-friendly interface, and also steers the application of the trained classi ers to data.