TY - GEN
T1 - Visualizing the trustworthiness of a projection
AU - Aupetit, Michaël
N1 - Publisher Copyright:
© 2006 i6doc.com publication. All rights reserved.
PY - 2006
Y1 - 2006
N2 - The visualization of continuous multi-dimensional data based on their projection in a 2-dimensional space is a way to detect visually interesting patterns, as far as the projection provides a faithful image of the original data. We propose to visualize directly in the projection space, how much the neighborhood has been preserved or not during the projection. We color the Vorono'i cells associated with the segments of the Delaunay graph of the projections, according to their stretching or compression. We experiment these techniques with the Principal Component Analysis and the Curvilinear Component Analysis applied to different databases.
AB - The visualization of continuous multi-dimensional data based on their projection in a 2-dimensional space is a way to detect visually interesting patterns, as far as the projection provides a faithful image of the original data. We propose to visualize directly in the projection space, how much the neighborhood has been preserved or not during the projection. We color the Vorono'i cells associated with the segments of the Delaunay graph of the projections, according to their stretching or compression. We experiment these techniques with the Principal Component Analysis and the Curvilinear Component Analysis applied to different databases.
UR - http://www.scopus.com/inward/record.url?scp=47349095471&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:47349095471
T3 - ESANN 2006 Proceedings - European Symposium on Artificial Neural Networks
SP - 271
EP - 276
BT - ESANN 2006 Proceedings - European Symposium on Artificial Neural Networks
PB - d-side publication
T2 - 14th European Symposium on Artificial Neural Networks, ESANN 2006
Y2 - 26 April 2006 through 28 April 2006
ER -