Results of Principal Component Analyses on a Database of Gallium Arsenide-based Zero Dimensional Nanostructures

The present work is based on a database, which has been created by us on III-V compound semiconductor-based droplet-epitaxially grown zero dimensional nanostructures. Principal component analysis (PCA) has been carried out on different sets of variables. PCA is best illustrated by matching a multidimensional ellipsoid onto the observation “cloud”. Generally, the magnitudes of projections of principal components onto parameter axes reflect well the importance of the given variable. Kohonen Self-Organizing Mapping was carried out on the projection of 9D data onto the plane of the first two principal components. Clustering have been observed.