• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.43.2023.tde-29052023-221306
Document
Author
Full name
Amanda Farias dos Santos
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2023
Supervisor
Committee
Abdalla, Elcio (President)
Oliveira, Claudia Lucia Mendes de
Souza, Carlos Alexandre Wuensche de
Title in Portuguese
Análise de dados fotométricos obtidos através do aprendizado de máquina e da K-d Tree
Keywords in Portuguese
aprendizado de máquina
K-d Tree
Redshift fotométrico
Abstract in Portuguese
Nesta dissertação utilizamos três modelos de aprendizados de máquina (machine learning) para o cálculo do redshift fotométrico de galáxias do catálogo astronômico do Dark Energy Survey (DES). Dois destes modelos são códigos públicos e nós construímos o terceiro modelo a partir de uma interface de programação de deep learning chamada keras. Para o treinamento dos modelos de machine learning, foram utilizadas as informações espectroscópicas das galáxias do catálogo astronômico do VIMOS Public Extragalactic Redshift Survey (VIPERS). Com o intuito de verificar a acurácia dos redshifts fotométricos, foi desenvolvida uma estrutura de dados chamada K-d Tree que separa as galáxias em subconjuntos de acordo com os seus dados fotométricos. Para cada subconjunto foi criado um outro aprendizado de máquina que calcula o quão preciso é o valor do redshift fotométrico calculado pelos três modelos para cada galáxia. Através deste resultado, foi possível excluir galáxias cujo redshift fotométrico está longe do valor do redshift espectroscópico.
Title in English
Analysis of photometric data obtained through machine learning and K-d Tree
Keywords in English
K-d Tree
machine learning
Photometric redshift
Abstract in English
In this thesis, we use three machine learning models to evaluate the photometric redshift of galaxies from the Dark Energy Survey (DES). Two of these models are open source and we built the third model using a deep learning programming interface called keras. In order to train the machine learning models, we used the spectroscopic information of galaxies from the astronomical catalog of the VIMOS Public Extragalactic Redshift Survey (VIPERS). In order to verify the accuracy of photometric redshifts, we developed a data structure called K-d Tree that separates galaxies into subsets according to their photometric data. In each subset, another machine learning was created. It evaluates how accurately is the photometric value of the redshift calculated by the three models for each galaxy is. Through this result, it was possible to exclude galaxies whose photometric redshift is far from the spectroscopic redshift value.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2023-07-17
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.