Feature Informativeness, Curse-of-Dimensionality and Error Probability in Discriminant Analysis
Abstract: This thesis is based on four papers on high-dimensional discriminant analysis. Throughout, the curse-of-dimensionality effect on the precision of the discrimination performance is emphasized. A growing dimension asymptotic approach is used for assessing this effect and the limiting error probability are taken as the performance criteria. A combined effect of a high dimensionality and feature informativeness on the discrimination performance is evaluated.
This dissertation MIGHT be available in PDF-format. Check this page to see if it is available for download.