Extracting information from interval data using symbolic principal component analysis

Maria do Rosário De Oliveira Silva; Pacheco A.; Valadas R.

Key information

Authors:

Maria do Rosário De Oliveira Silva (Maria do Rosário De Oliveira Silva); Pacheco A. (António Manuel Pacheco Pires); Valadas R. (Rui Jorge Morais Tomaz Valadas)

Published in

01/01/2017

Abstract

<jats:p>We introduce generic definitions of symbolic variance and covariance for random interval-valued variables, that lead to a unified and insightful interpretation of four known symbolic principal component estimation methods: CPCA, VPCA, CIPCA, and SymCovPCA. Moreover, we propose the use of truncated versions of symbolic principal components, that use a strict subset of the original symbolic variables, as a way to improve the interpretation of symbolic principal components. Furthermore, the analysis of a real dataset leads to a meaningful characterization of Internet traffic applications, while highligting similarities between the symbolic principal component estimation methods considered in the paper.</jats:p>