Predicting RPC efficiency using machine learning and deep learning models

The report discusses main steps of retrieving RPC (Resistive Plate Chambers) data of multiple LHC (Large Hadron Collider) runs. The retrieved data contains different metrics that are being analyzed by the help of the main statistical methods. Once the data becomes familiar, different Machine Learning...

Descripción completa

Detalles Bibliográficos
Autor principal: Michelevicius, Mikas
Lenguaje:eng
Publicado: 2021
Materias:
Mathematical Physics and Mathematics
Computing and Computers
Acceso en línea:http://cds.cern.ch/record/2784121
_version_ 1780972087002791936
author Michelevicius, Mikas
author_facet Michelevicius, Mikas
author_sort Michelevicius, Mikas
collection CERN
description The report discusses main steps of retrieving RPC (Resistive Plate Chambers) data of multiple LHC (Large Hadron Collider) runs. The retrieved data contains different metrics that are being analyzed by the help of the main statistical methods. Once the data becomes familiar, different Machine Learning and Deep Learning models are trained to predict the fiducial efficiency of each chamber based on the measurable detector quantities collected. Data of CMS experiment of 2018 is used to train the models. For the work, Python scikit-learn and Keras libraries are used to train the models. The predicted results are then compared against the initial fiducial efficiency values
id cern-2784121
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2021
record_format invenio
spelling cern-27841212021-10-14T21:03:44Zhttp://cds.cern.ch/record/2784121engMichelevicius, MikasPredicting RPC efficiency using machine learning and deep learning modelsMathematical Physics and MathematicsComputing and ComputersThe report discusses main steps of retrieving RPC (Resistive Plate Chambers) data of multiple LHC (Large Hadron Collider) runs. The retrieved data contains different metrics that are being analyzed by the help of the main statistical methods. Once the data becomes familiar, different Machine Learning and Deep Learning models are trained to predict the fiducial efficiency of each chamber based on the measurable detector quantities collected. Data of CMS experiment of 2018 is used to train the models. For the work, Python scikit-learn and Keras libraries are used to train the models. The predicted results are then compared against the initial fiducial efficiency valuesCERN-STUDENTS-Note-2021-204oai:cds.cern.ch:27841212021-10-14
spellingShingle Mathematical Physics and Mathematics
Computing and Computers
Michelevicius, Mikas
Predicting RPC efficiency using machine learning and deep learning models
title Predicting RPC efficiency using machine learning and deep learning models
title_full Predicting RPC efficiency using machine learning and deep learning models
title_fullStr Predicting RPC efficiency using machine learning and deep learning models
title_full_unstemmed Predicting RPC efficiency using machine learning and deep learning models
title_short Predicting RPC efficiency using machine learning and deep learning models
title_sort predicting rpc efficiency using machine learning and deep learning models
topic Mathematical Physics and Mathematics
Computing and Computers
url http://cds.cern.ch/record/2784121
work_keys_str_mv AT micheleviciusmikas predictingrpcefficiencyusingmachinelearninganddeeplearningmodels

Ejemplares similares