A Study on the Influence of Shape in Classifying Small Spectral Data Sets

Classification of spectral data has raised a growing interest in may research areas. However, this type of data usually suffers from the curse of dimensionality. This causes most statistical methods and/or classifiers to not perform well. A recently proposed alternative which can help avoiding this problem is the Dissimilarity Representation, in which objects are represented by their dissimilarities to representative objects of each class. However, this approach depends on the selection of a suitable dissimilarity measure. For spectra, the incorporation of information on their shape, can be significant for a good discrimination. In this paper, we make a study on the benefit of using a measure which takes shape of spectra into account. We show that the shape-based measure not only leads to better classification results, but that a certain number of objects is enough to achieve it. The experiments are conducted on three one-dimensional data sets and a two-dimensional one.