Measurement of Higgs boson production via vector boson fusion in the H → WW → 2l2v channel at 13 TeV using deep neural networks

• Ano de defesa: 2019
• Autor(a) principal: Rosas, Luis Junior Sánchez
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade do Estado do Rio de Janeiro; Centro de Tecnologia e Ciências::Instituto de Física Armando Dias Tavares; BR; UERJ; Programa de Pós-Graduação em Física
• Palavras-chave em Português: Higgs; VBF; DNN; CMS/CERN; CNPQ::CIENCIAS EXATAS E DA TERRA::FISICA::FISICA DAS PARTICULAS ELEMENTARES E CAMPOS
• Link de acesso: http://www.bdtd.uerj.br/handle/1/12824

A study of standard model Higgs physics of elementary particles approaching gauge invariance, spontaneous symmetry breaking, proton-proton interactions, phenomenology, and experimental Higgs boson highlights. The CMS is a multipurpose detector of the LHC/CERN. Operating a superconducting solenoid and includes dedicated subsystems for tracking of charged particles close to point of interaction, measurements of electromagnetic and hadronic energy deposits and muon tracking out of the solenoid. The CMS Trigger and Data Acquisition System efficiently reduces the event rate to about 1 kHz for permanent storage and analysis off-line. Object definition, reconstruction, efficiency and systematics used in the H → WW analysis for the full dataset are presented. Electrons, muons, jets, and missing transverse moments are basic elements of event reconstruction for analysis. The measurement of single and double lepton efficiencies is fulfilled to be used in the analysis. An analysis of the Higgs boson production via vector boson fusion in the H → WW → 2l2v(l = e, µ) channel. The analysis strategy, the control regions, the datadriven estimates and the results are also presented. The data of pp collision used correspond to the integrated luminosity of 35.9 fb−1 collected during 2016 period at √s = 13 TeV. The channel has a final state with 2 isolated leptons of high transverse momentum and a high value of missing energy transverse, because neutrinos escape detection. Events are classified em categories according to the number of jets with high pT in the final state and the VBF analysis, cut-based, has only one category of 2 jets. Deep neural networks are used for optimal discrimination between VBF signal and main standard model background processes, for the final signal extraction. The final result of the analysis is obtained using two selection processes: a cut-based analysis and another based on the shape distribution of a one-dimensional discriminant created with the output of the neural network.