This page contains my lectures given at the BaBar Analysis School at the SLAC National Laboratory, Stanford, California October 25-30 2009: (school web page).

• Lecture 1: Introduction, Cut based analyses, Fisher Discriminant, Neural Networks lecture1.pdf
• Lecture 2: Decision trees, TMVA, and a recap of what to do with your analysis. lecture2.pdf
• Lecture 3: Tutorial using data from BaBar: lecture3.pdf, example tar file, lecture3-results.pdf.

These lectures provide an introduction to some of the techniques used to discriminate between two or more classes of events, such as the classic signal and background problem often encountered in experimental particle physics. The techniques covered include a treatment of cut-based analyses, Fisher discriminants, Artificial Neural Networks and Decision trees. Having covered the basic concepts required to understand these methods in lectures 1 and 2, there is an example session that uses TMVA and a sample of BaBar data to develop and compare several different MVAs that can be used to distinguish between signal and background samples. This example illustrates some of the issues related to incorporating an MVA into the subsequent part of the data analysis chain that is often required before one is able to attain the desired end result.

Back to the Teaching home-page