Nonlinear Discriminative Data Visualization

Apr, 2009

Due to the tremendous increase of electronic information with respect to the size of data sets as well as dimensionality, visualization of high-dimensional data constitutes one of the key problems of data mining. Since embedding in lower dimensions necessarily includes a loss of information, methods to explicitly control the information kept by a specific visualization technique are highly desirable. The incorporation of supervised class information constitutes an important specific case. In this contribution we propose an extension of prototype-based local matrix learning by a charting technique which results in an efficient nonlinear dimension reduction and discriminative visualization of a given labelled data manifold

Link

@INPROCEEDINGS{Bunte_ESANN2009a,
author = {Kerstin Bunte and Barbara Hammer and Petra Schneider and Michael Biehl},
title = {{Nonlinear Discriminative Data Visualization}},
booktitle = {Proc. of the  17th "European Symposium on Artificial Neural Networks (ESANN)},
month = {"Apr."},
pages = {65--70},
editor = {M. Verleysen},
publisher = {D-facto Publications},
address = {Bruges, Belgium},
year = {2009},
url = {http://hdl.handle.net/11370/2ef21297-4bb5-45bf-b756-0edc3a9b5239},
url2 = {http://www.cs.rug.nl/~biehl/Preprints/discr_visesann09.pdf},
abstract = {Due to the tremendous increase of electronic information with respect to the size of data sets as well as dimensionality,  visualization of high-dimensional data constitutes one of the key problems of data mining.  Since embedding in lower dimensions necessarily includes a loss of information, methods to explicitly control the information kept by a specific visualization technique are highly desirable.  The incorporation of supervised class information constitutes an important specific case.  In this contribution we propose an extension of prototype-based local matrix learning by a charting technique which results in an efficient nonlinear dimension reduction  and discriminative visualization of a given labelled data manifold},
}