Informatics and Applications
2021, Volume 15, Issue 4, pp 72-78
ON THE CHOICE OF PARTIAL ORDERS ON FEATURE VALUES SETS IN THE SUPERVISED CLASSIFICATION PROBLEM
- E. V. Djukova
- G. O. Masliakov
Abstract
The authors consider one of the central problems of machine learning - the supervised classification.
A scheme for the logical classification algorithms synthesis is described under the assumption that the features descriptions of precedents are the elements of the finite partial orders Cartesian product. A criterion for the correctness of the voting algorithm of representative elementary classifiers is formulated. The authors study the possibility of defining linear orders on sets of feature values that provide better classification, which is not necessarily correct, in assumption that the source data are not ordered (the precedents descriptions are the elements of the antichains product). A procedure is proposed for "correct" consistent ordering of the acceptable values of separate features, while the remaining features are antichains. The results of experiments on real data are presented demonstrating the effectiveness of the methods developed in the work.
[+] References (9)
- Baskakova, L., and Yu. Zhuravlev. 1981. Amodel of recognition algorithms with representative sampls an systems of
supportingsets. USSRComp. Math. Math. Phys. 21(5):189-199.
- Djukova, E., and N. Peskov. 2002. Search for informative fragments of object descriptions in discrete recognition
procedures. Comp. Math. Math. Phys. 42(5):711-723.
- Zhuravlev, Yu. I., Y.V. Ryazanov, and O.Y. Sen'ko. 2006.
Raspoznavanie. Matematicheskie metody. Programmnaya
sistema. Primeneniya [Recognition. Mathematical methods. Software system. Applications]. Moscow: FAZIS.
176 p.
- Djukova, E., G. Maslyakov, and P. Prokofyev. 2019. On
the logical analysis of partially ordered data in the supervised classification problem. Comp. Math. Math. Phys. 59(9):1542-1552.
- Janostik, R., J. Konecny, and P. Kraj ca. 2020. Interface between logical analysis of data and formal concept analysis.
Eur. J. Oper. Res. 284(2):792-800.
- Djukova, E., G. Maslyakov, and P. Prokofjev. 2018. Dualization problem over the product of chains: Asymptotic
estimates for the number of solutions. Dokl. Math. 98:564-567.
- Baklanova, A., E. Djukova, and G. Masliakov. 2020. Investigation of the dependence of the supervised classification
quality on the choice of partial orders on feature values
sets. Intelligent data processing: Theory and applications. Moscow: Russian Academy of Sciences. 24–26.
- Sotnezov, R. 2009. Genetic algorithms for problems of
logical data analysis in discrete optimization and image
recognition. Pattern Recognit.Image Anal. 19(3):469–477.
- Tarjan, R. 1976. Edge-disjoint spanning trees and depth-first search //Acta Inform. 6(2):171–185.
[+] About this article
Title
ON THE CHOICE OF PARTIAL ORDERS ON FEATURE VALUES SETS IN THE SUPERVISED CLASSIFICATION PROBLEM
Journal
Informatics and Applications
2021, Volume 15, Issue 4, pp 72-78
Cover Date
2021-12-30
DOI
10.14357/19922264210410
Print ISSN
1992-2264
Publisher
Institute of Informatics Problems, Russian Academy of Sciences
Additional Links
Key words
machine learning; logical classification algorithms; correct supervised classification algorithm; partially ordered set; Cartesian product of partial orders; linear order; dualization over product of partial orders
Authors
E. V. Djukova  and G. O. Masliakov
Author Affiliations
Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
|