The methodology for selecting a programming language is proposed to develop support tools for managing the educational process in a regional university, with focus on soft models and calculations. The study shows that it is necessary to provide additional stability for a regional university in “turbulent” conditions in the case of a significant number of non-factors affecting both the educational process itself and its management. The study presents an implementation of the method claimed. It is proposed to use the following programming languages under conditions of uncertainty: DELPHI, Visual Basic, and Python, the most common one in modern applications.

The study deals with obtaining numerical estimates of shogi pieces’ disposition on the board with the help of software. Using these estimates, the authors intend to evaluate a player’s skill. That requires collecting data, reviewing and repeating existing approaches, proposing and implementing the authors’ approach, and comparing the results. The difference in strength of two players is calculated based on their win distribution via resimulation of the Ferreira approach. In implementing Yamashita’s approach, the relation between Elo rating and the average degradation of the position estimate as a result of a bad move is searched for. The authors’ approach is proposed to estimate a player’s skill. Hence, the dependence of wins, good and bad moves, the deviation of the estimates of real and ideal moves, and the average improvement of the position from the Elo rating are analyzed. The study attempts to cluster plays and moves.

A simulation model of a customer service system in the bank based on agent and discrete-event modeling is developed, analyzed and described. The study presents the results of experiments aimed at analyzing the performance of the customer service system in the bank. The statistics of queues and average service time at different working hours are studied. The results obtained can be used for economic planning and decision-making to improve the bank’s performance. The materials discussed in the article can also be used to provide students with the necessary professional skills when studying for bachelor’s and master’s degrees.

The article studies and evaluates the chaotic dynamics parameters of complex non-stationary biosystems, in particular, a specific chaos type occurring in the parameters that describe the systems mentioned. Contrary to Lawrence deterministic chaos, these systems provide neither the process of repetition nor the repetition of the system’s initial state. The study is based on the evaluation of the degree of chaotic time series of cardiac signals of a human cardiovascular system using methods of non-parametric mathematical statistics and those of non-linear dynamics. On the example of elements of cardiac signals (time series cardiac intervals) of a cardiovascular system and with autocorrelation analysis, the study shows that biosystem parameters have a strictly chaotic structure and non-linear trend. The chaotic processes in cardiac performance are demonstrated by computing the largest Lyapunov exponent and, as a result, a con-stant sign (+, −) change in each cardiac signal recording. The study can be used to identify parameters of the functional state of a human organism.

The article deals with a new software for automated data collection. Using parsing of open websites, this software allows obtaining a set of the most significant financial and economic indicators of firms for a particular industry in the region in a short time and presents it in an Excel document, resulting in a comprehensive financial and economic analysis of a given industry in the region. A list of required firms published at export-base.ru is a necessary condition for data collection. Python and such key libraries as request, selenium, beautifulsoup4, and csv are selected as the main tool for writing the program. The software developed is tested for quickly obtaining complex data from IT companies in the Rostov Oblast to conduct the financial and economic analysis of the IT industry in the region.

Neural network models are a universal tool for solving probabilistic problems in various fields and areas. However, when solving a set of problems, typical neural networks and training algorithms can lead to overfitted networks, resulting in a decrease in the accuracy of the posterior distribution of the output layer of the neural network. In this case, the network fails to adapt fully to self-training due to no data in the training sample, leading to a decrease in the network’s ability to detect anomalies. One of the methods to solve the problem is by using Bayesian methods for training of and inference in neural networks. When using the Bayesian approach in neural networks, it is intended to set the weights in the network as a probability distribution over all values that are permissible for the parameters of the network. Then, classical learning mechanisms and probabilistic inference used in Bayesian networks, such as algorithms based on the Monte Carlo methods and Markov chains, can be used to determine weights. The article considers the issues of using Bayesian neural networks for modeling the process of web application testing.

The article considers the issues of selecting key parameters for the diagnosis of liver disease. The analytic hierarchy process is used to determine the importance of criteria based on the assessment of the applied medicine provided by the experts. The selection of key parameters outline for the diagnosis of liver disease is built. Matrices of pair-wise comparisons are obtained to determine the significance of each introduced criterion. When filling in the matrix of pair-wise comparisons, the experts proposed their assessments, and their contradiction have been calculated with the consistency index. The comparative analysis of parameters according to the obtained values of global priorities with Pearson correlation values is carried out. In the course of the study, parameters with the most effective correlation of criteria are selected. Those will help reduce the study area of interrelation of selected parameters, as well as liver disease stages.

Decision-making methods based on probability theory, specific chapters in mathematics, and other tools dictate the necessity to update approaches and methods for aquiring new competencies, and even to return to intuitive decision-making. Decision-making for a person (a decision maker) is becoming a more complex objective, particularly in medicine, as there is no certainty in the formalization of tasks, with high responsibility for the decision made. The article considers the design and implementation of a decision-making support system made as a module of a medical information system to assist cardiovascular surgeons in solving problems related to clinical decision-making in phlebology.

The article studies the flame propagation in a soap bubble filled with a spherical volume of a homogeneous pre-mixed combustible gas ignited by an electrical discharge. Visualization of the flame front propagation is carried out using a high-speed digital video camera. The video recordings of an experiment series show the relationship between the flame front velocity and the time and radial values of a soap bubble in various conditions.
The experiment results can be used for the invention of a modern combustion chamber and more sophisticated burner arrangements.

The article studies the process of using fixed specified convolutional kernels. The calculations are conducted considering the change in filter size and stride. In the course of the study, four different sets of images and structures of neural networks are selected. The article discusses four types of specified kernels: vertical, horizontal, diagonal, and ring. The dependency of the accuracy of image recognition in monochrome and color is analyzed. A positive shift is observed when training fixed kernels.

The article considers geometric models of target evasion from a pursuer moving in a straight line and uniformly. In geometric models, it is supposed that a pursuer has a detection area. The study discusses the following detection models: circular, sector, and angular. Initially, the target is moving out of the line of sight along the optimal trajectory to leave the area of probable detection. The issues of the minimum distance from a pursuer allowing the target to escape the detection area are analyzed. It is noted that the adaptive behavior of the pursuer is not simulated. As a result of the study, animated images have been produced to present the process of target evasion from the pursuer.

The temperature effect on the rotatory power of optically active substances when using light sources with different wavelengths is determined experimentally for each case. The commonly known data is provided for λ = 589.3 nm. The article determines the temperature dependence of the specific rotation of sugar solution when using a He-Ne laser with a wavelength of λ = 632.8 nm. The experiments are carried out at the temperature range of 20–80 °C for various solution concentrations. The recorded dependence is linear with an angular coefficient of 0.043 degrees℃. As the temperature increases, the value of the specific rotation decreases slightly.

The article describes a software package aimed at solving the problem of searching for pairs of prime numbers equidistant from a certain arbitrary number. The results of work with the selected numbers 231 and 232 are obtained at the limit of the capabilities of mass-produced personal computers. Two different algorithms for solving the problem have been developed using the Python programming language and are described herein. The results’ agreement obtained via two independent algorithms can be interpreted as verification of the problem solution results, which are difficult to identify visually due to large numerical values. The obtained pairs of prime numbers can be used as bases in modular arithmetic designed to work with large numbers, with bivalent effect eliminated.