Statistical Learning: the impact of combining feature-information structures on total genetic characterization of composite data points

摘要

In this paper, we evaluate the impact of combining two or more feature spaces on the total genetic characterization of composite data points. In this context, a composite data point is a segmented genome of influenza viral (i.e. a dataset of biosequences). The outcomes show different estimate (i.e trees) for the information structure of composite data points. This approach can be used in phylogenetic analysis to predict the tree-structure (total genetic characterization) of composite data points using combined feature spaces.