Abstract
To ensure the effective functioning of the university educational environment, document flow processes automation, which includes the task of documents automatic classification, is of great ...importance. The article considers the task of classifying university documents by machine learning methods in order to improve the quality of classification. Documents preprocessing was carried out, which made it possible to distinguish significant words in documents, due to which the accuracy of documents classification increased. Described are methods of extracting features from text TF and TF-IDF, which determine keywords by words frequency included in document. A modification of the TF-IDF method is proposed, which consists in calculating the words importance depending on their part of speech. This made it possible to improve the classification quality by highlighting only important and significant words in documents. Suggested is a classification algorithm using a method of support vectors to reduce the documents number involved in classification and a method of k-nearest neighbor for classification. The advantage of this algorithm over the described analogues is shown, which is expressed in the number of mistakenly classified documents decrease.
Abstract
The article considers an approach to the processing and analysis of biomedical data from clinical examination. An automated diagnosis system is presented to be applied in liver disease ...diagnosis by the results of the initial examination of patients using a classifier based on fuzzy inference. The research is carried out by modeling the proposed system in the environment MATLAB FuzzyLogic. The regularities are revealed making it possible to normalize the results of laboratory tests depending on gender characteristics and to increase the percentage of correctly determined disease stages.
Abstract
The article deals with optimization issues in design of an automatic control system (ACS). The method of multicriteria optimization is presented, which combines simulation of ACS and ...optimization procedure based on a genetic algorithm and implemented by the MATLAB / Global Optimization Toolbox. The method of the generalized criterion composition based on the initial particular criteria is also proposed. The use of this generalized criterion makes it possible to reduce the complexity of determining the optimal parameters of the ACS and allows to take into account the constraints when searching for the optimal solution.
MONITORING OF REINTRODUCED RARE PLANTS USING UAV DATA
The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences,
12/2023
Conference Proceeding
Abstract
The method of analytical modeling and simulation of technological processes is presented, which makes it possible to assess the influence of external factors on the operation of information ...and control systems. The application of this method to study the process of detecting pipeline defects by vibroacoustic control systems is considered. A mathematical model with automatic selection of the parameters of autocorrelation functions of random disturbances is proposed, which makes it possible to take into account the influence of natural and man-made noises on the detection process. The possibility of modeling the influence of random disturbances on the working processes of technological machines is also considered. The proposed simulation models are of practical value, since they allow one to study virtual prototypes of construction and road-building machinery in various modes and reduce the cost of performing expensive field experiments.
Alluaudite phases are very attractive as both cathode and electrolyte materials for rechargeable sodium-ion batteries. In this work, the combined experimental and DFT studies have been performed to ...establish the diffusion mechanism in alluaudite-like compound Na5In(MoO4)4. The ionic conductivity was found to reach 3.3 × 10−4 S/сm at 687 K, with an activation energy of 0.66 eV. The sodium diffusion mechanisms have been revealed from the analysis of the 23Na NMR spectra along with the DFT estimations of Na-ion migration barriers. Our results predict that one-dimensional diffusion of sodium in the separate channels along the c-axis is accompanied by the cross-linking jumps providing two-dimensional diffusion in the bc-plane. It is clearly demonstrated that the indium deficiency favors 2D diffusion, but sharply increases the energy barrier for 1D diffusion. Comparison of our results for Na5In(MoO4)4 and related Na5Sc(MoO4)4 shows that the type and deficiency of M-metal in the NaxMy(MoO4)z alluaudites can control the sodium diffusion. The present work highlights the key aspects of cation influence on the diffusion properties in alluaudite materials.
The simulation and optimization technology is proposed for creating control systems that are designed for heat and power facilities. This paper presents an automatic feed control system of steam ...generators in the nuclear power plant unit on the basis of microprocessor controllers. The technology of simulation and optimization combines the methods of statistical modelling of disturbance signals, the control system simulation and the controller parameters optimization using genetic algorithm. As the initial data for the model design, we used archival information about processes on the existing unit equipment. The results of simulation studies and optimization of automatic control system's parameters are derived in MATLAB.
The saturation magnetizations, local magnetic anisotropy field and coercivity values of FeCo film plated under various processing conditions have been investigated to optimize soft magnetic ...properties. Herein, we introduce a very promising processing technique based on the electrodeless deposition of FeCo film with carbohydrates as reducing agents. The produced FeCo film demonstrated significantly better saturation magnetization values and less contaminations, compared to those for the sample preparing with conventional reducing agent (sodium hypophosphite). The surface morphology and coercivities of FeCo films are dependent on the iron content and type of reducing agent. The coercive force values range from 12 up to 30 Oe and the saturation magnetization from 150 to 240 emu/g depending on the bath composition and deposition parameters. Maximum of saturation magnetization magnitude is reached for FeCo film with 30% cobalt. The local anisotropy field value of FeCo alloys increases with a decrease in Fe content for all reducing agent types.
Macroscopic and local magnetic anisotropy of CoFeB-SiO2/α-Si:H60 multilayer films was studied in comparison with that for the thick CoFeB-SiO2 granular films. The volume fraction of magnetic ...component Co40Fe40B20 is ranged from 0.34 to 0.70. In the multilayer films the magnetic layer and silicon spacer thicknesses are 2.5÷4.5nm and 3.5nm correspondingly. The magnetization curves and ferromagnetic resonance measurements were used to determining magnetic anisotropy energy constants. The main contribution to the local magnetic anisotropy energy constant is found to be from surface magnetic anisotropy of magnetic granules. It is found that the magnetic percolation point of multilayer films (XC~40) is shifted in comparison with that in granular films (XC~30). Surface magnetic anisotropy energy constants of multilayer films with different X evaluated from the surface modes of spin-wave resonance are in the range 0.5<Ks <0.9erg/cm2.