The main goals of the present research are: a) to provide an alternative method for reducing the inherent dimensionality of hypercubes and simultaneously preserving the informational content of data; b) to use the reduced dataset in order to develop a classification method based on spectral analysis; c) to demonstrate the benefits of the present method versus the cluster analysis of the original dataset. In order to achieve these purposes, we used a number of simulated hypercubes and evaluated CLUEGO performances versus one of the most common clustering algorithms (e.g., the k-means) applied to both the original and reduced data cubes.
CLUEGO, AN INFORMATIONAL HYPERSPECTRAL CLASSIFIER.
POMPILIO, Loredana;MARINANGELI, Lucia
2013-01-01
Abstract
The main goals of the present research are: a) to provide an alternative method for reducing the inherent dimensionality of hypercubes and simultaneously preserving the informational content of data; b) to use the reduced dataset in order to develop a classification method based on spectral analysis; c) to demonstrate the benefits of the present method versus the cluster analysis of the original dataset. In order to achieve these purposes, we used a number of simulated hypercubes and evaluated CLUEGO performances versus one of the most common clustering algorithms (e.g., the k-means) applied to both the original and reduced data cubes.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
