HiggsSearch Masterclass (pupils and/or students)

Brief overview

This is a collection of materials that can be used as a template for tasks that can be worked on in the master class ( or even over several days) by pupils or by students (the more advanced version). The sections are factorized and can be adapted and extended according to the students' level of knowledge. Motivation for this collection is the discovery of the Higgs Bosons in 2012, therefore the published data collections from the CMS detector [1] from 2012 (8 TeV data set) and the publication of the 13 TeV data sets from Atlas [2] are used.The aim of this introduction is to give the students a transparent overview of the discovery of the Higgs boson in the four lepton channel and to motivate the statistical reasoning about the actual discovery with the help of Python, that is also fundamentally presented.

Language used

• German
• English

Execution viability

It is possible to perform the repo using MyBinder. Some of the notebooks rely on graphical applications that require opening in a new window. There exists a _remote_ variant of these, which should be used instead. For the local application, both options are available (the implementation of the individual graphical applications differs slightly due to the different functional scope of the packages used, but contains all important features). However, it is recommended to use the _remote variants, as they fit better into the notebook environment. If this is the first time that Python is used, it is recommended to have a look at the notebook containing Python basics too.

• To run the repo remotely:
  

• To run locally (following commands for Terminal/Power Shell):

  git clone https://github.com/a-monsch/HiggsSearch__P4_and_Masterclass
  cd HiggsSearch__P4_and_Masterclass
  

  With an optional virtual environment:

  virtualenv venv_higgs
  # Linux
  source venv_higgs/bin/activate
  # Windows
  .\venv\Scripts\activate
  

  The necessary Python (>= 3.6) packages are listed below (pip install <package>) but can also be downloaded automatically via

  pip3 install -r binder/requirements.txt
  

  • SciPy
  • NumPy
  • Pandas
  • matplotlib
  • kafe2 (current master) (optional)
  • iminuit
  • PyQt5 (if running locally)
  • Jupyter
  • tqdm
  • jupyter contrib nbextensions (optional)

  If the virtual environment is used, a kernel for the jupyter notebooks should be reregistered.

  ipython kernel install --user --name=venv_higgs
  

  The jupyteter notebook can be started directly from the root directory of the repository with

  jupyter notebook
  

  After shutting down the notebook, you can leave the virtual environment with deactivate.

Provided datasets

In the MyBinder version the data sets have been downloaded automatically. For the local version, the data records can either be downloaded here manually and unpacked (several times) in a folder data or be downloaded and unpacked automatically by sh /binder/postBuild (Linux). Included in the record are following folders:

• for_event_display

  Contains the .ig files for different decays that can be loaded in IspyWebGL [3] (is called in the notebooks). With the help of the values taken there and possible combinations of the detailed tables (also in the folder), the invariant mass of individual decay (four lepton decay) events can be calculated. (Detailed information about the origin of the event files can be found in README_EVENTS.md in the folder)

• for_widgets

  Contains lists of the Monte Carlo (MC) simulated four lepton invariant masses of the background process and Higgs bosons of different masses (signal process). Is used to display the underlying background and individual signal MCs in the graphical application (in combination with for_event_display calculated masses)

• for_longer_analysis

  Contains strongly reduced data sets of the CMS detector from 2012 [1] (Run B and C, more details) to be processed in the context of the 'for_student' notebook. Also included is the parameterization of the signal process for different masses using a Gaussian distribution (for the application of the Likelihood Ratio Test). Finally, a summarized list of four lepton invariant masses from the ATLAS measurement, as well as the bin entries of the histograms of the simulated background and signal process are included [2] (more details).

References

[1] Bin Anuar Afiq Aizuddin Jomhari Nur Zulaiha Geiser Achim. Higgs-to-four-leptonanalysis example using 2011-2012 data. 2017. DOI: 10.7483/OPENDATA.CMS.JKB8.D634. URL: http://opendata.vern/record/5500.

[2] ATLAS Collaboration. ATLAS 13 TeV samples collection at least four leptons (electron or muon), for 2020 Open Data release. URL: http://opendata.cern.ch/record/15005.

[3] T McCauley. “A browser-based event display for the CMS Experiment at the LHCusing WebGL”. In:Journal of Physics: Conference Series898 (Okt. 2017), S. 072030. DOI: 10.1088/1742-6596/898/7/072030. URL: https://doi.org/10.1088%2F1742-6596%2F898%2F7%2F072030.