|
INFORMATICS
Completeness criteria for a linear model of classification algorithms with respect to families of decision rules
A. G. Dyakonova, A. M. Golovinab a Lomonosov Moscow State University, Moscow, Russian Federation
b Bauman Moscow State Technical University, Moscow, Russian Federation
Abstract:
In the framework of Zhuravlev’s algebraic approach to classification problems, a linear model of algorithms is investigated (estimates of class membership are generated by linear regressions). The possibility of weakening the completeness requirement (obtaining an arbitrary estimation matrix) in order to obtain any classification of a fixed set of objects by using special decision rules is investigated.
Keywords:
classification, linear algorithms, algebraic approach, correctness, completeness.
Citation:
A. G. Dyakonov, A. M. Golovina, “Completeness criteria for a linear model of classification algorithms with respect to families of decision rules”, Dokl. RAN. Math. Inf. Proc. Upr., 490 (2020), 67–70; Dokl. Math., 101:1 (2020), 57–59
Linking options:
https://www.mathnet.ru/eng/danma36 https://www.mathnet.ru/eng/danma/v490/p67
|
Statistics & downloads: |
Abstract page: | 102 | Full-text PDF : | 45 | References: | 20 |
|