• português (Brasil)
    • English
    • español
  • English 
    • português (Brasil)
    • English
    • español
  • Login
About
  • Policies
  • Instructions to authors
  • Contact
    • Policies
    • Instructions to authors
    • Contact
View Item 
  •   Home
  • Centro de Ciências em Gestão e Tecnologia - CCGT
  • Programas de Pós-Graduação
  • Ciência da Computação - PPGCC-So
  • Teses e dissertações
  • View Item
  •   Home
  • Centro de Ciências em Gestão e Tecnologia - CCGT
  • Programas de Pós-Graduação
  • Ciência da Computação - PPGCC-So
  • Teses e dissertações
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Browse

All of DSpaceCommunities & CollectionsBy Issue DateAuthorsAdvisorTitlesSubjectsCNPq SubjectsGraduate ProgramDocument TypeThis CollectionBy Issue DateAuthorsAdvisorTitlesSubjectsCNPq SubjectsGraduate ProgramDocument Type

My Account

Login

Aprendiz de descritores de mistura gaussiana

Thumbnail
View/Open
tese.pdf (2.186Mb)
carta comprovante (155.6Kb)
Date
2017-12-14
Author
Freitas, Breno Lima de
Metadata
Show full item record
Abstract
For the last decades, many Machine Learning methods have been proposed aiming categorizing data. Given many tentative models, those methods try to find the one that fits the dataset by building a hypothesis that predicts unseen samples reasonably well. One of the main concerns in that regard is selecting a model that performs well in unseen samples not overfitting on the known data. In this work, we introduce a classification method based on the minimum description length principle, which naturally offers a tradeoff between model complexity and data fit. The proposed method is multiclass, online and is generic in the regard of data representation. The experiments conducted in real datasets with many different characteristics, have shown that the proposed method is statiscally equivalent to the other classical baseline methods in the literature in the offline scenario and it performed better than some when tested in an online scenario. Moreover, the method has proven to be robust to overfitting and data normalization which poses great features a classifier must have in order to deal with large, complex and real-world classification problems.
URI
https://repositorio.ufscar.br/handle/ufscar/9249
Collections
  • Teses e dissertações

UFSCar
Universidade Federal de São Carlos - UFSCar
Send Feedback

UFSCar

IBICT
 

 


UFSCar
Universidade Federal de São Carlos - UFSCar
Send Feedback

UFSCar

IBICT