Empirical data modeling in terms of mixtures of normal
distributions and kernel density estimation was thus shown to
lead to detailed description of features pertaining to the data
set at hand, pinpointing technological aspects deserving
further attention aimed at meeting exacting specifications in a
cost-effective way.