• 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.55.2016.tde-16112016-170837
Document
Author
Full name
Bruno Kim Medeiros Cesar
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Carlos, 2016
Supervisor
Committee
Monaco, Francisco José (President)
Delbem, Alexandre Cláudio Botazzo
Fernandes, Marcio Merino
Title in Portuguese
Estudo e extensão da metodologia DAMICORE para tarefas de classificação
Keywords in Portuguese
Aprendizado de máquina
Classificação
DAMICORE
Dissertação
Abstract in Portuguese
A área de aprendizado de máquina adquiriu grande importância na última década graças à sua capacidade de analisar conjuntos de dados complexos em larga escala. Em diversas áreas do conhecimento existe a demanda pela análise de dados por especialistas, seja para obter agrupamentos latentes ou classificar instâncias em classes conhecidas. As ferramentas acessíveis a especialistas leigos em programação são limitadas a problemas específicos e demandam um custo de desenvolvimento às vezes proibitivo, sendo interessante buscar por ferramentas genéricas e aplicáveis a qualquer área do conhecimento. Este trabalho busca estender e implementar uma metodologia genérica de aprendizado de máquina capaz de analisar quaisquer conjuntos de arquivos de forma praticamente livre de configuração. Foram obtidos resultados satisfatórios de sua aplicação em um conjunto amplo de problemas para agrupamento e classificação de executáveis, spam e detecção de línguas.
Title in English
Research and extension of the DAMICORE methodology for classification tasks
Keywords in English
Classification
DAMICORE
Machine learning
Masters thesis
Abstract in English
Machine learning has rised in importance in the last decade thanks to its power to analyse complex datasets in large scale. At several areas of knowledge there is a demand for data analysis by domain experts, be it for discovering latent clusters or classifying instances into known groups. The tools available for experts that do not master computer programming are limited to specific tasks and demand a high development cost, which sometimes is prohibitive. It is interesting, then, to develop generic tools useful to any area of knowledge. This masters thesis seeks to extend and implement a generic machine learning methodology capable of analysing any set of files mostly free of configuration. Its application produced satisfactory results in a wide set of clustering and classification problems over binary executables, spam classification, and language identification.
 
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
2016-11-16
 
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.