Geodesic
Estimation of the reliability of a classification algorithm as based on a new information model
Žurnal vyčislitelʹnoj matematiki i matematičeskoj fiziki, Tome 53 (2013) no. 5, pp. 808-815 
S. I. Gurov. Estimation of the reliability of a classification algorithm as based on a new information model. Žurnal vyčislitelʹnoj matematiki i matematičeskoj fiziki, Tome 53 (2013) no. 5, pp. 808-815. http://geodesic.mathdoc.fr/item/ZVMMF_2013_53_5_a11/
@article{ZVMMF_2013_53_5_a11,
     author = {S. I. Gurov},
     title = {Estimation of the reliability of a classification algorithm as based on a new information model},
     journal = {\v{Z}urnal vy\v{c}islitelʹnoj matematiki i matemati\v{c}eskoj fiziki},
     pages = {808--815},
     year = {2013},
     volume = {53},
     number = {5},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/ZVMMF_2013_53_5_a11/}
}
TY  - JOUR
AU  - S. I. Gurov
TI  - Estimation of the reliability of a classification algorithm as based on a new information model
JO  - Žurnal vyčislitelʹnoj matematiki i matematičeskoj fiziki
PY  - 2013
SP  - 808
EP  - 815
VL  - 53
IS  - 5
UR  - http://geodesic.mathdoc.fr/item/ZVMMF_2013_53_5_a11/
LA  - ru
ID  - ZVMMF_2013_53_5_a11
ER  - 
%0 Journal Article
%A S. I. Gurov
%T Estimation of the reliability of a classification algorithm as based on a new information model
%J Žurnal vyčislitelʹnoj matematiki i matematičeskoj fiziki
%D 2013
%P 808-815
%V 53
%N 5
%U http://geodesic.mathdoc.fr/item/ZVMMF_2013_53_5_a11/
%G ru
%F ZVMMF_2013_53_5_a11

Voir la notice de l'article provenant de la source Math-Net.Ru

A new approach to the estimation of the reliability of classification algorithms is proposed. The approach is based on an unconventional information model of such algorithms. Examples of new estimates are given, which are compared with usual statistical estimates.

[1] Vorontsov K. V., “Obzor sovremennykh issledovanii po probleme kachestva obucheniya algoritmov”, Tavrich. vestn. informatiki i matem., 2004, no. 1, 5–24 | MR | Zbl

[2] Gurov S. I., “Informatsionnaya model algoritma klassifikatsii i otsenka ego nadezhnosti”, Intellektualizatsiya obrabotki informatsii, 8-ya mezhdunar. nauchn. konf. IOI-8, Sb. dokl. (Kipr, g. Pafos, 17–23 oktyabrya 2010 g.), MAKS Press, M., 2010

[3] Vorontsov K. V., Kombinatornaya teoriya nadezhnosti obucheniya po pretsedentam, Dis ... dok. fiz.-matem. nauk. 05-13-17, VTs RAN, 2010, 271 pp. http://www.machinelearning.ru/wiki/images/b/b6/Voron10doct.pdf

[4] Lbov G. S., Startseva N. G., Logicheskie reshayuschie funktsii i voprosy statisticheskoi ustoichivosti reshenii, Izd-vo In-ta matem., Novosibirsk, 1999 | MR | Zbl

[5] Gurov S. I., Otsenka nadezhnosti klassifitsiruyuschikh algoritmov, MGU VMK, M., 2002

[6] Itskhoki O., “Vybor modeli i paradoksy prognozirovaniya”, Kvantil, 2006, no. 1, 43–51

[7] Gurov S. I., “Otsenka veroyatnosti $0$-sobytiya”, Vestn. Tverskogo gos. un-ta. Ser. Prikl. matem., 2009, no. 14, 55–66

[8] Bolshev L. N., Smirnov N. V., Tablitsy matematicheskoi statistiki, Nauka, M., 1965 | MR

[9] Zaks L., Statisticheskoe otsenivanie, Statistika, M., 1976 | MR

[10] Kendal M., Styuart A., Statisticheskie vyvody i svyazi, Nauka, M., 1973 | Zbl

[11] Uilks S., Matematicheskaya statistika, Nauka, M., 1967 | MR

[12] Gurov S. I., “Intervalnoe otsenivanie na osnove printsipa soglasovannosti”, Vestn. Tverskogo gos. un-ta. Ser. Prikl. matem., 2008, no. 9, 77–93