Search
 
loader loading subjects...
cover
Cartogram-Based Data Visualization Using the Growing Hierarchical SOM
Àngela Martín, Alfredo Vellido
249 - 252
10.3233/978-1-61499-320-9-249
Frontiers in Artificial Intelligence and Applications
Volume 256: Artificial Intelligence Research and Development
Abstract

Model interpretability is a problem of knowledge extraction from the patterns found in data to which data visualization can contribute. Nonlinear dimensionality reduction techniques provide flexible visual insight, but their locally varying representation distortion makes interpretation far from intuitive. In this paper, we apply a cartogram method, based on techniques of geographic representation, to data visualization. It allows reintroducing this distortion, measured as a U-matrix, in the visual maps of the Growing Hierarchical Self Organising Map (GHSOM).

$35.00 / €27.50 / £22.00
Add PDF to cart