Master's Dissertation
DOI
https://doi.org/10.11606/D.55.2003.tde-26092003-122130
Document
Author
Full name
Jose Fernando Rodrigues Junior
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Carlos, 2003
Supervisor
Committee
Traina, Agma Juci Machado (President)
Morimoto, Carlos Hitoshi
Oliveira, Maria Cristina Ferreira de
Title in Portuguese
"Desenvolvimento de um Framework para Análise Visual de Informações Suportando Data Mining"
Keywords in Portuguese
Descoberta de Conhecimento em Bases de Dados
Mineração de Dados
Visualização de Informações
Abstract in Portuguese
No presente documento são reunidas as colaborações de inúmeros trabalhos das áreas de
Bancos de Dados, Descoberta de Conhecimento em Bases de Dados, Mineração de Dados,
e Visualização de Informações Auxiliada por Computador que, juntos, estruturam o tema de
pesquisa e trabalho da dissertação de Mestrado: a Visualização de Informações. A teoria
relevante é revista e relacionada para dar suporte às atividades conclusivas teóricas e práticas
relatadas no trabalho.
O referido trabalho, embasado pela substância teórica pesquisada, faz diversas
contribuições à ciência em voga, a Visualização de Informações, apresentando-as através de
propostas formalizadas no decorrer deste texto e através de resultados práticos na forma de
softwares habilitados à exploração visual de informações. As idéias apresentadas se baseiam
na exibição visual de análises numéricas estatísticas básicas, frequenciais (Frequency Plot),
e de relevância (Relevance Plot). São relatadas também as contribuições à ferramenta
FastMapDB do Grupo de Bases de Dados e Imagens do ICMC-USP em conjunto com os
resultados de sua utilização. Ainda, é apresentado o Arcabouço, previsto no projeto original,
para construção de ferramentas visuais de análise, sua arquitetura, características e utilização.
Por fim, é descrito o Pipeline de visualização decorrente da junção entre o Arcabouço de
visualização e a ferramenta FastMapDB.
O trabalho se encerra com uma breve análise da ciência de Visualização de
Informações com base na literatura estudada, sendo traçado um cenário do estado da arte
desta disciplina com sugestões de futuros trabalhos.
Title in English
"Development of a Framework for Visual Analysis of Information with Data Mining suport"
Keywords in English
Data Mining
Information Visualization
Knowledge Discovery in Databases
Abstract in English
In the present document are joined the collaborations of many works from the fields of
Databases, Knowledge Discovery in Databases, Data Mining, and Computer-based
Information Visualization, collaborations that, together, define the structure of the research
theme and the work of the Masters Dissertation presented herein. This research topic is the
Information Visualization discipline, and its relevant theory is reviewed and related to support
the concluding activities, both theoretical and practical, reported in this work.
The referred work, anchored by the theoretical substance that was studied, makes
several contributions to the science in investigation, the Information Visualization, presenting
them through formalized proposals described across this text, and through practical results
in the form of software enabled to the visual exploration of information. The presented ideas
are based on the visual exhibition of numeric analysis, named basic statistics, frequency
analysis (Frequency Plot), and according to a relevance analysis (Relevance Plot). There are
also reported the contributions to the FastMapDB tool, a visual exploration tool built by the
Grupo de Bases de Dados e Imagens do ICMC-USP, the performed enhancements are listed
as achieved results in the text. Also, it is presented the Framework, as previewed in this work's
original proposal, projected to allow the construction of visual analysis tools; besides its
description are listed its architecture, characteristics and utilization. At last, it is described the
visualization Pipeline that emerges from the joining of the visualization Framework and the
FastMapDB tool.
The work ends with a brief analysis of the Information Visualization science based on
the studied literature, it is delineated a scenario of the state of the art of this discipline along
with suggestions for future work.
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
2004-07-29
WARNING: The material described below relates to works resulting from this thesis or dissertation. The contents of these works are the author's responsibility.
- RODRIGUES, J.F., et al. Using efficient visual exploration techniques to evaluate features for content-based image retrieval [doi:10.1109/SIBGRA.2003.1241007]. In 16th Brazilian Symposium on Computer Graphics and Image Processing (SIBGRAPI 2003) [online], Sao Carlos, Brazil, . Sao Carlos, Brazil : IEEE Comput. Soc, 2003. p. 183-190.
- RODRIGUES, J.F., TRAINA, A.J.M., and TRAINA, C.. Enhanced visual evaluation of feature extractors for image mining [doi:10.1109/AICCSA.2005.1387042]. In The 3rd ACS/IEEE International Conference onComputer Systems and Applications, 2005. [online], Cairo, Egypt, . Cairo, Egypt : IEEE, 2005. p. 201-208.
- RODRIGUES, J.F., TRAINA, A.J.M., and TRAINA, C.. Frequency plot and relevance plot to enhance visual data exploration [doi:10.1109/SIBGRA.2003.1240999]. In 16th Brazilian Symposium on Computer Graphics and Image Processing (SIBGRAPI 2003) [online], Sao Carlos, Brazil, . Sao Carlos, Brazil : IEEE Comput. Soc, 2003. p. 117-124.