• 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.55.2003.tde-03122014-084211
Document
Author
Full name
Eduardo Jose Tejada Gamero
Institute/School/College
Knowledge Area
Date of Defense
Published
São Carlos, 2003
Supervisor
Committee
Minghim, Rosane (President)
Oliveira, João Batista Souza de
Traina, Agma Juci Machado
Title in Portuguese
Agrupamento visual em grandes conjuntos de dados multidimensionais
Keywords in Portuguese
Não disponível
Abstract in Portuguese
O agrupamento visual de dados é uma abordagem que integra conceitos de visualização de informação e de aprendizado de máquina, especificamente por agrupamento, em um único algoritmo. Isso permite aproveitar a capacidade dos seres humanos para tomar decisões baseados em conhecimento específico de domínio, bem como a capacidade dos computadores para registrar, armazenar, manipular, e recuperar dados. Este trabalho apresenta a pesquisa realizada nessa linha, a qual tratou problemas específicos presentes nos métodos de agrupamento visual de dados. O principal problema em estudo foi a escalabilidade de algoritmos de agrupamento visual baseados em densidade, propondo meios para acelerar os processos de cálculo de projeções e de estimativa de densidade envolvidos neles. Como resultado é apresentado o algoritmo HC-Enhanced, que possui desempenho bastante superior ao algoritmo HC-Cooperative no qual é baseado, além de um dispositivo de melhoria do agrupamento.
Title in English
Visual Clustering for large multidimensional data sets
Keywords in English
Not available
Abstract in English
Visual clustering is an approach that integrates concepts from information visualization and machine learning, specifically clustering, in a single algorithm. This allows combine the capability of humans beings to make decisions based on their domain knowledge, and the capability of the computers for registering, storing, manipulating, and retrieving data. This work presents the research done on this line, which involved treating specific problems visual clustering approaches present. The main problem studied was the scalability of density-based visual clustering algorithms, proposing means for accelerating the projection and density estimation processes involved in them. As a result it is presented the algorithm HC-Enhanced, whose performance is considerably improved compared to the algorithm HC-Cooperative on which it is based, as well as a component for clustering quality improvement.
 
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
2014-12-03
 
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.