The article considers a solution to a two-class recognition problem using metric al gorithms based on k-nearest neighbors. The proposed solution options are: the classical KNN algorithm based on a nonparametric estimate of the probability density function; the Nadaraya-Watson NW method for calculating the regression dependence based on the determined values of the target feature; calculating estimates in the local neighbor hood of objects and dividing their values into non-overlapping intervals with dominance of representatives of one of the two classes. The optimality criteria were the majority rule for KNN, the accuracy of dividing the values of the target feature on the numerical axis for NW and the stability of object estimates within the interval boundaries. When searching for the optimal k values for the listed options, the sliding control method was used. The results of the obtained solutions are demonstrated on the GERMAN sample from the repository based on the descriptions of 1000 objects in a heterogeneous feature space. The efficiency of choosing the k values based on the stability of object estim.
| Mualliflar | Ignatev , N.A., Tursunmurotov, D.Kh., Игнатьев , Н.А., Турсунмуротов, Д.Х. |
|---|---|
| Jurnal | Ҳисоблаш ва амалий математика муаммолари |
| Nashr sanasi | 2024-10-11 |
| Son | 4 |
| Betlar | 117-125 |
| Til | Rus |
Nadaraya-Watson method, decision rule, objective function, feature stability, Метод Надарая– Ватсона, решающее правило, целевая функция, устойчивость признака
The article analyzes the work done so far to create a high-quality computer translation. It also describes the approaches, challenges, and solutions in achieving semantically accurate translation in computer…
The paper proposes a method that belongs to the group of intensional methods of pattern recognition, the essence of which is the transition from a system of initial features to a reduced system of models of…
Now in the world there are a lot of diseases threatening humanity and for these medical treatments and diagnostic methods have also improved both in practical and statistical. This paper discusses one of the most…
A numerical study was conducted based on the proposed mathematical apparatus of non-stationary technology for filtering a low-concentration solution through a porous medium, taking into account the process of…
The article discusses modeling the quality management system of educational information systems (EIS) using fuzzy cognitive maps, as one of the most effective approaches to solving weakly formalized problems. To…
In this article, within the framework of a mathematical model of three-phase filtration of incompressible and immiscible liquids and gas without taking into account the difference between phase pressures and…
Regimes of stochasticity and chaos are allowed in nonlinear reaction-diffusion systems that describe chemical-reactive, astrophysical, thermal, thermonuclear, plasma, epidemiological and demographic processes. The work…
Based on the developed mathematical software, this paper presents the results of numerical calculations on the task of monitoring and forecasting the concentration of industrial emissions of pollutants (CO2, dust…
In the modeling of a number of practical problems, it is brought to the mixed problems imposed on the system of quasi-linear hyperbolic equations. Since there is no exact solution to such problems, they are solved using…
Today it is much more difficult to imagine life without computers. There is not a single area that a computer cannot penetrate. Everyday life is rapidly digitizing and unfortunately, more and more people among us are…