PH4100 MSci Major Projects - ATLAS ttH (H-bb)

Introduction

This project is based on the analysis of simulated data of proton-proton collisions in the ATLAS particle detector at the Large Hadron Collider (LHC). The objective is to study the production of Higgs particles in association with a pair of top quarks: the so-called ttH process. Specifically the process of interest (the signal) is when the Higgs boson decays to a pair of b quarks: ttH (H-bb). The project will required the detailed analysis of simulated samples of signal events, as well as analysing all the relevant background processes. In particular, the student will aim to develop event selection procedures to effectively separate the signal from the backgrounds. This will start with the development of a simple cut-based event selection, optimised using relevant metrics. The student can then progress to using instead an event selection based on eg a Fisher discriminant, or more advanced multivariate methods (such as a Neural Network (NN) or a Boosted Decision Tree (BDT)).

Physics: important sources of information

Published articles:

Normally it would be expected that students doing this project are taking the PH4515 Statistical Data Analysis course taught by Prof. Glen Cowan.

The Particle Data Group (PDG) website includes a lot of useful information. In particular, the student should use the "Summary Tables" to find out the main decay modes and respective branching ratios for the following particles: top, Higgs, W and Z. The decays of the tau lepton to electrons or muons (as opposed to hadronic decays of the tau lepton) are also of interest for this project.

The LHC Higgs cross-section working group TWiki pages are here.

The cross-sections of most background processes, as measured by the ATLAS experiment, are available from these plots: Standard Model Summary Plots (July 2019)

Tools for the project

This is a computer-based project. You will be analysing large simulated samples of events in the ATLAS detector at the LHC. The project requires programming in C++, the use of the ROOT data analysis package and knowledge of basic Linux OS commands. Links to sources of information on these and other topics are given in the sub-topics below.

C++ programming language

Students doing this project are expected to have some experience with programming already, and ideally have prior knowledge and experience of the C++ programming language (such as that gained from taking the PH3170 C++ and Object Oriented Programming course).

Links to some useful C++ resources:

  • The PH4515 Statistical Data Analysis course includes revision of C++: slides available here.
  • The RHUL library has many different books on C++ programming.
  • A C++ online reference guide, with tutorials and more: www.cplusplus.com.
  • Another C++ online reference guide: cppreference.com.

ROOT data analysis package

  • The homepage of ROOT at CERN has much useful information, including the User Guide for ROOT 6 in html or as a single pdf file.
  • A first look at a signal ntuple: see instructions here.
  • Using TSelector framework to process ntuples
    • In this project you will analyse ntuples containing variables defined for each event. In ROOT, ntuples are called "Trees" and the variables are sometimes referred to as the "leafs".
    • See the Chapter "Trees" in the ROOT User Guide for more information, and especially the sub-section "Using TTree::MakeSelector" which describes the TSelector framework that is used in this project.
  • Using TLorentz four-vectors: see the Chapter "Physics Vectors", and especially the sub-section that describes four-vectors in ROOT ("TLorentzVectors").

A possible part of this project is to implement the event selection using multivariate methods. For more information on this see this separate page.

Linux OS

The RHUL Particle Physics computer cluster (linappserv) uses the Linux operating system. An introduction to this system(Linux/Unix), including most of the commands you'll need for this project is given in slides 4-10 of this pdf file. An introduction to the most commonly used Linux commands is also available here.

Editing files (using the emacs or vi editors)

In this project you will be writing/editing C++ code very often. You can in principle use any ASCII text editor for this, but it is recommended that you use an editor that is C++ aware. One such editor (happens to be my favourite) is the Gnu Emacs editor. If you decide to use emacs here is a handy reference card with the main commands; you'll probably only ever need about 10 commands or so. A simpler editor (which, like emacs, is also available on the RHUL Particle Physics linux computer cluster) is vi. There are many available online introductions to the vi editor: here is one.

How to connect to the linappserv linux computer cluster from your laptop or from a PC in the Teaching Lab

The most up-to-date instructions for this are normally on the web page of the PH4515 Statistical Data Analysis course (taught by Prof. Glen Cowan). The direct link to the instructions is here.

Troubleshooting:

  • If connecting at RHUL, it seems that being connected to the CampusNet network (rather than eg eduroam) is important.
  • On the Teaching Lab PCs, you may have to download PuTTY if it is not already installed. If that is the case find and download the respective the MS Installer file (*.msi) on the web. After the download is complete the file should auto-install.

Simulated event samples

The list of simulated event samples available for this project is given in a separate page.

(Information on an older list of simulated samples is still available in this page, just for reference.)

Analysis framework and workflow

The following diagram is a schematic depiction of the basic framework and work flow for the analysis.

schematic

The student will be given access to all the required simulated data samples as well as template code files (eg analysis_ttH_PH4100.C and myplots_display_PH4100.C) to start with. The latter will need to be added to by the student, as they develop their own data analysis. The list of files required to run the analysis code is

 run_analysis_PH4100.C
 analysis_ttH_PH4100_v3.0-FOR-STUDENT.C
 analysis_ttH_PH4100.h
 functions/prot_makeCutFlowTables.C
 functions/func_makeCutFlowTables-PH4100mod.C
 functions/prot_identifyProcess_082019Prod.C
 functions/func_identifyProcess_082019Prod.C

The student will also need the following root program to prepare publication-quality plots for their report:

myplots_display_PH4100_v2.2.C 

-- Pedro Teixeira Dias - 20 Jan 2021

Comments

Edit | Attach | Watch | Print version | History: r20 < r19 < r18 < r17 < r16 | Backlinks | Raw View | Raw edit | More topic actions

Physics WebpagesRHUL WebpagesCampus Connect • Royal Holloway, University of London, Egham, Surrey TW20 0EX; Tel/Fax +44 (0)1784 434455/437520

Topic revision: r20 - 28 Oct 2024 - PedroTeixeiraDias

 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2024 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding RHUL Physics Department TWiki? Send feedback