AN EVALUATION OF INTRINSIC DIMENSIONALITY ESTIMATORS

The intrinsic dimensionality of a data set may be useful for understan ding the properties of classifiers applied to it and thereby for the s election of an optimal classifier. In this paper we compare the algori thms for two estimators of the intrinsic dimensionality of a given dat a set and extend their capabilities. One algorithm is based on the loc al eigenvalues of the covariance matrix in several small regions in th e feature space. The other estimates the intrinsic dimensionality from the distribution of the distances from an arbitrary data vector to a selection of its neighbors. The characteristics of the two estimators are investigated and the results are compared. It is found that both c an be applied successfully, but that they might fail in certain cases. The estimators are compared and illustrated using data generated from chromosome banding profiles.