Abstract
The problem of choosing an attribute space for solving classification problems is considered. When choosing a space, nonlinear transformations of the values of both individual features and their combinations are used. The fuzziness of dividing objects into classes according to the values of individual features is investigated using membership functions. The generality of the procedure for calculating the values of the membership function for data measured in nominal and interval scales is shown. The efficiency of the feature space selection is demonstrated through the indicators of the generalizing ability of the algorithms calculated through the estimates of the structure of relations of class objects.
Similar content being viewed by others
REFERENCES
I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning (MIT Press, 2016; DMK Press, Moscow, 2018).
https://towardsdatascience.com/ensemble-methods-bagging-boosting-and-stacking-c9214a10a205.
N. A. Ignatyev, “Structure choice for relations between objects in metric classification algorithms,” Pattern Recognit. Image Anal. 28, 590–597 (2018).
N. A. Ignat’ev, “Computing generalized parameters and data mining,” Autom. Remote Control 72, 1068 (2011).
D. Y. Saidov, “Data visualization and its proof by compactness criterion of objects of classes,” Int. J. Intell. Syst. Appl. 9 (8), 51–58 (2017).
A. Asuncion and D. J. Newman, UCI Machine Learning Repository (Univ. Calif., Irvine, 2007). https://www.ics.uci.edu/~mlearn/MLRepository.html.
Funding
The work was carried out within the framework of the research program of the Department of Algorithms and Programming Technologies of the National University of Uzbekistan.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
The author declare he has no conflict of interests.
Additional information
Nikolai Alexandrovich Ignatiev. Born in 1950. He graduated from Tashkent State University in 1976 with a degree in applied mathematics. Candidate of Physical and Mathematical Sciences (1988), Doctor of Physical and Mathematical Sciences (2005). Professor of the Department of Algorithms and Programming Technologies of the National University of Uzbekistan. Research interests: mathematical modeling and data mining methods. Published over 90 scientific articles and three monographs.
Rights and permissions
About this article
Cite this article
Ignatiev, N.A. On Nonlinear Transformations of Features Based on the Functions of Objects Belonging to Classes. Pattern Recognit. Image Anal. 31, 197–204 (2021). https://doi.org/10.1134/S1054661821020085
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S1054661821020085