Generación de métricas en análisis de datos

In this work we introduce primary generators of metrics, that consist of linearcombinations of known metrics. We establish some theoretical properties of these generators,related to the computation of principal components, principal correlations andinertia (generalized variance). We apply primary ge...

Descripción completa

Autores Principales: Bolaños Guerrero, Fabricio, Trejos Zelaya, Javier
Formato: Artículo
Idioma: Español
Publicado: 2015
Acceso en línea: http://revistas.ucr.ac.cr/index.php/matematica/article/view/218
http://hdl.handle.net/10669/12860
Sumario: In this work we introduce primary generators of metrics, that consist of linearcombinations of known metrics. We establish some theoretical properties of these generators,related to the computation of principal components, principal correlations andinertia (generalized variance). We apply primary generators in Principal ComponentAnalysis using some combinations of classic metrics.Keywords: Principal Component Analysis, Euclidean metrics, correlation.