Interactive Construction of Classification Trees Using Treemaps
Manoel Mendonça (1), Christiane de Costa Santana (1), Daniela Soares Cruzes (1)
e-mails: mgmn@unifacs.br, christiane.santana@unifacs.br, daniela@unifacs.br
(1) Universidade Salvador - Nuperc 40171-100 Salvador Brasil
Most of the approaches published in the literature proposes
a completely automatic process to generate decision trees.
These approaches miss valuable expert tacit knowledge input
during the construction of the tree. This paper describes an
approach for interactive construction of decision trees. The
approach is user-centered. It combines the strengths of the
user and the computer to build better decision trees. The
user provides domain knowledge and evaluates intermediate
results of the algorithm. The computer automatically creates
patterns satisfying user constraints and generates
appropriate visualizations of the produced tree. A tool was
developed to support this approach. It combines treemap
visualization, visual data mining mechanisms, and the J48
(Weka) algorithm to interactively build a decision tree.
Keywords:Visual Data Mining, Decision Tree, Classification
BibTex
@INPROCEEDINGS{mendonça04:173,
AUTHOR = {Manoel Mendonça and Christiane de Costa Santana and Daniela Soares Cruzes},
TITLE = {Interactive Construction of Classification Trees Using Treemaps},
BOOKTITLE = {30ma Conferencia Latinoamericana de Informática (CLEI2004)},
YEAR = {2004},
editor = {Mauricio Solar and David Fernández-Baca and Ernesto Cuadros-Vargas},
pages = {576--585},
address = {},
month = Sep,
organization = {Sociedad Peruana de Computación},
note = {ISBN 9972-9876-2-0},
file = {http://clei2004.spc.org.pe/es/html/pdfs/173.pdf}
}
PDF de este artículo
PDF de CLEI2004 (incluye todos los artículos)
Página principal CLEI 2004
Generado por Sociedad Peruana de Computación