论文信息 - Dimension Reduction for Visual Data Mining

Dimension Reduction for Visual Data Mining

We present a method for dimension reduction applied to visual data mining in order to reduce the user cognitive load due to the density of data to be visualized and mined. We use consensus theory to address this problem: the decision of a committee of experts (in our case existing attribute selection methods) is generally better than the decision of a single expert. We illustrate the choices operated for our algorithm and we explain the results. We compare successfully these results with those of two widely used methods in attribute selection, a filter based method (LVF) and a wrapper based method (Stepclass).

Edwige P. Fangseu Badjio | François Poulet | F. Poulet | Edwige Fangseu Badjio

[1] Ron Kohavi,et al. Wrappers for Feature Subset Selection , 1997, Artif. Intell..

[2] Catherine Blake,et al. UCI Repository of machine learning databases , 1998 .

[3] Haim Levkowitz,et al. From Visual Data Exploration to Visual Data Mining: A Survey , 2003, IEEE Trans. Vis. Comput. Graph..

[4] Mark A. Hall,et al. Correlation-based Feature Selection for Discrete and Numeric Class Machine Learning , 1999, ICML.

[5] Gregory Piatetsky-Shapiro,et al. Advances in Knowledge Discovery and Data Mining , 2004, Lecture Notes in Computer Science.

[6] Colin Ware,et al. Information Visualization: Perception for Design , 2000 .

[7] Susan T. Dumais,et al. Inductive learning algorithms and representations for text categorization , 1998, CIKM '98.

[8] Mihael Ankerst,et al. Visual Data Mining , 2001, Encyclopedia of GIS.

[9] François Poulet,et al. Cooperation between automatic algorithms , interactive algorithms and visualization tools for Visual Data Mining , 2002 .

[10] Yoshua Bengio,et al. Pattern Recognition and Neural Networks , 1995 .

[11] Huan Liu,et al. A Probabilistic Approach to Feature Selection - A Filter Solution , 1996, ICML.

[12] John M. Chambers,et al. Graphical Methods for Data Analysis , 1983 .