• 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.10.2021.tde-08112022-155912
Document
Author
Full name
Renata Martins de Carvalho
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2020
Supervisor
Committee
Baquero, Oswaldo Santos (President)
Barbosa, David Soeiro
Guilloux, Aline Gil Alves
Title in Portuguese
Modelagem preditiva da presença e número de gatos nos domicílios brasileiros
Keywords in Portuguese
Aprendizado de máquina
Epidemiologia veterinária
Felinos
Abstract in Portuguese
Os gatos domésticos estão cada vez mais presentes nos domicílios, superando os cães em número em diferentes países. A escolha do gato como animal de companhia pode envolver diferentes aspectos, e é possível que características demográficas, socioeconômicas e geográficas das populações humanas estejam associadas à presença e ao número desses animais nos lares. No presente estudo, os algoritmos de aprendizado de máquina supervisionado Partial Least Square, Random Forest e Extreme Gradient Boosting foram aplicados aos dados demográficos e socioeconômicos provenientes da Pesquisa Nacional de Saúde de 2013 para identificar preditores da presença e número de gatos nos domicílios brasileiros e classificá-los de acordo com sua contribuição ao desempenho preditivo dos modelos construídos. Dentre eles, destacaram-se o número de cães, a zona de localização do domicílio (urbana ou rural) e número de moradores maiores de 18 anos. Porém, os algoritmos só explicaram uma pequena fração da complexidade que determina a coabitação entre humanos e gatos, mesmo incorporando 47 preditores socioeconômicos, geográficos e demográficos da população humana. Pesquisas qualitativas podem identificar preditores mais relevantes e informar estudos preditivos de base populacional.
Title in English
Predictive modeling of presence and number of cats in Brazilian households
Keywords in English
Feline
Machine learning
Veterinary epidemiology
Abstract in English
Domestic cats′ presence in households is increasing, outnumbering dogs in different countries. The choice of keeping a cat as a pet might involve many aspects, and the demographic, socioeconomic, and geographical characteristics of human populations might be associated with the presence and number of those animals in residences. In the present study, we predicted the presence and number of cats in Brazilian households and classified predictors according to their predictive performance. To this end, we used data from the 2013 National Health Survey and three supervised machine learning algorithms: Partial Least Square, Random Forest, and Extreme Gradient Boosting. The number of dogs, the household zone (urban or rural), and the number of residents over 18 years old had the highest predictive performance. However, the algorithms explained only a small fraction of the complexity determining human-cat cohabitation, even with 47 socioeconomic, geographic, and demographic predictors of the human population. Qualitative research might identify more relevant predictors and inform population-based predictive studies.
 
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
2022-12-27
 
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.