Geodesic
On the optimality of the max-depth and max-rank classifiers for spherical data
Applications of Mathematics, Tome 65 (2020) no. 3, pp. 331-342.

Voir la notice de l'article provenant de la source Czech Digital Mathematics Library

The main goal of supervised learning is to construct a function from labeled training data which assigns arbitrary new data points to one of the labels. Classification tasks may be solved by using some measures of data point centrality with respect to the labeled groups considered. Such a measure of centrality is called data depth. In this paper, we investigate conditions under which depth-based classifiers for directional data are optimal. We show that such classifiers are equivalent to the Bayes (optimal) classifier when the considered distributions are rotationally symmetric, unimodal, differ only in location and have equal priors. The necessity of such assumptions is also discussed.
DOI : 10.21136/AM.2020.0331-19
Classification : 62G30, 62H30
Keywords: depth-based classifier; von Mises-Fisher distribution; directional data; cosine depth 
@article{10_21136_AM_2020_0331_19,
     author = {Venc\'alek, Ond\v{r}ej and Demni, Houyem and Messaoud, Amor and Porzio, Giovanni C.},
     title = {On the optimality of the max-depth and max-rank classifiers for spherical data},
     journal = {Applications of Mathematics},
     pages = {331--342},
     publisher = {mathdoc},
     volume = {65},
     number = {3},
     year = {2020},
     doi = {10.21136/AM.2020.0331-19},
     mrnumber = {4114256},
     language = {en},
     url = {http://geodesic.mathdoc.fr/articles/10.21136/AM.2020.0331-19/}
}
TY  - JOUR
AU  - Vencálek, Ondřej
AU  - Demni, Houyem
AU  - Messaoud, Amor
AU  - Porzio, Giovanni C.
TI  - On the optimality of the max-depth and max-rank classifiers for spherical data
JO  - Applications of Mathematics
PY  - 2020
SP  - 331
EP  - 342
VL  - 65
IS  - 3
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/articles/10.21136/AM.2020.0331-19/
DO  - 10.21136/AM.2020.0331-19
LA  - en
ID  - 10_21136_AM_2020_0331_19
ER  - 
%0 Journal Article
%A Vencálek, Ondřej
%A Demni, Houyem
%A Messaoud, Amor
%A Porzio, Giovanni C.
%T On the optimality of the max-depth and max-rank classifiers for spherical data
%J Applications of Mathematics
%D 2020
%P 331-342
%V 65
%N 3
%I mathdoc
%U http://geodesic.mathdoc.fr/articles/10.21136/AM.2020.0331-19/
%R 10.21136/AM.2020.0331-19
%G en
%F 10_21136_AM_2020_0331_19
Vencálek, Ondřej; Demni, Houyem; Messaoud, Amor; Porzio, Giovanni C. On the optimality of the max-depth and max-rank classifiers for spherical data. Applications of Mathematics, Tome 65 (2020) no. 3, pp. 331-342. doi : 10.21136/AM.2020.0331-19. http://geodesic.mathdoc.fr/articles/10.21136/AM.2020.0331-19/

Cité par Sources :