Abstract:
A classification model under partial information in the form of expectations of features is proposed. It is based on the minimax and minimin decision strategies. The discriminant function is computed by maximizing (minimizing) the risk functional as a measure of classification error over a set of probability distributions with bounds determined by the information about features, and its minimizing over a set of parameters. An algorithm is reduced to the parametric linear programming.
Keywords:
classification, machine learning, linear programming, risk functional, loss function, expectation, minimax strategy.
Citation:
L. V. Utkin, Yu. A. Zhuk, I. A. Selihovkin, “A classification model on the basis of partial information about features in the form of their mean values”, Artificial Intelligence and Decision Making, 2012, no. 2, 16–26; Scientific and Technical Information Processing, 39:6 (2012), 336–344
\Bibitem{UtkZhuSel12}
\by L.~V.~Utkin, Yu.~A.~Zhuk, I.~A.~Selihovkin
\paper A classification model on the basis of partial information about features in the form of their mean values
\jour Artificial Intelligence and Decision Making
\yr 2012
\issue 2
\pages 16--26
\mathnet{http://mi.mathnet.ru/iipr427}
\elib{https://elibrary.ru/item.asp?id=18941943}
\transl
\jour Scientific and Technical Information Processing
\yr 2012
\vol 39
\issue 6
\pages 336--344
\crossref{https://doi.org/10.3103/S0147688212060068}
This publication is cited in the following 3 articles:
V. I. Erokhin, A. P. Kadochnikov, S. V. Sotnikov, “Linear Binary Classification under Interval Uncertainty of Data”, Sci. Tech. Inf. Proc., 51:6 (2024), 539
A. Petrochenkov, I. Luzyanin, 2017 XX IEEE International Conference on Soft Computing and Measurements (SCM), 2017, 469
Konstantin N. Krasnykh, Anton B. Petrochenkov, Bernd Krause, 2016 XIX IEEE International Conference on Soft Computing and Measurements (SCM), 2016, 466