• 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.45.2023.tde-08022024-095259
Document
Author
Full name
Vinicius Santos Oliveira
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2023
Supervisor
Committee
Lima, Antonio Carlos Pedroso de (President)
Artes, Rinaldo
Ritter, Victor Silva
Title in Portuguese
Métodos de árvores de decisão em análise de sobrevivência: uma aplicação a dados de câncer
Keywords in Portuguese
Análise de sobrevivência
Árvore de decisão
Árvore de sobrevivência
Ensembles
Predição
Abstract in Portuguese
A análise de sobrevivência é um conjunto de técnicas estatísticas amplamente utilizadas para analisar tempos até a ocorrência de um ou mais eventos. Dentre dos possíveis métodos de modelagem preditiva para dados de sobrevivência, as árvores de decisão têm destaque devido à sua capacidade de modelar relações complexas entre as covaríáveis e a ocorrência do evento de interesse. Neste trabalho, são estudadas técnicas de árvore de decisão para dados censurados, revisando suas metodologias, avaliando suas vantagens e desvantangens e apresentando extensões com uso de ensembles. Por fim, as diferentes técnicas são aplicadas ao conjunto de dados do ICESP e comparadas com a abordagem usual baseada no modelo de riscos proporcionais de Cox usando métricas de avaliação de performance e técnicas de validação cruzada.
Title in English
Decision tree methods in survival analysis: an application to cancer dataset
Keywords in English
Decision tree
Ensembles
Prediction
Survival analysis
Survival tree
Abstract in English
Survival analysis is a set of statistical techniques widely used to analyze the time to the occurrence of one ore more events. Among the possible predictive modeling methods for survival data, decision trees stand out for their ability to model complex relationships between covariates and the occurrence of the event of interest. In this work, decision tree techniques for censored data are studied, their methods are reviewed, evaluating advantages and disadvantages, and their extensions using ensembles are presented. Finally, the different techniques are applied to the ICESP dataset and compared with the Cox proportional hazards model using predictive performance metrics and cross validation techniques.
 
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
2024-03-01
 
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.