Anatoliy Batyuk

Lviv Polytechnic National University · Department of Automated Control Systems

In the early software development stages, the aim of estimation is to obtain a rough understanding of the timeline and resources required to implement a potential project. The current study is devoted to a method of preliminary estimation applicable at the beginning of the software development life cycle when the level of uncertainty is high. The a...

Estimation is an essential step of software development project planning that has a significant impact on project success—underestimation often leads to problems with the delivery or even causes project failure. An important aspect that the classical estimation methods are usually missing is the Agile nature of development processes in the implemen...

The study is devoted to the development of information technology for forecasting based on time series. It has been found that it is important to develop new models and forecasting methods to improve the quality of the forecast. Information technology is based on the evolutionary method of synthesis of the forecast scheme grounded on basic forecast...

The paper deals with questions related to the ability of real-weighted bithreshold neurons and neural networks to solve the classification tasks. We study how many partitions of a finite set in n-dimensional vector space can be computed by using bithreshold neurons. The upper and lower bounds on the number of bithreshold dichotomies are proved. Our...

The paper deals with the issues concerning the application of bithreshold-like neural units in the solving of the classification tasks. The fully connected 2-layer feedforward network architecture with the hidden layer consisting of bithreshold neurons is considered in the paper. We design the neural network multicategory classifier based on such a...

The paper deals with the issues concerning the expressive power of bithreshold neurons and their application in pattern classification. First, we study how many Boolean functions of n variables can be computed by using bithreshold neural units. Our approach is based on the application of combinatorial methods in the theory of neural computation. Th...

The method of construction of the combined model of forecast ing of time series based on basic models of forecasting is developed in the work. The set of basic models is dynamic, ie new prediction models can be included in this set. Models also can be deleted depending on the properties of the time series. For the synthesis of a combined model of f...

The problem of the learning of polynomial threshold units over a fixed set of polynomials is treated in the paper. We consider two approaches in the supervised learning: off-line relaxation-like algorithms and spectral off-line algorithms. Our research is focused on answering the question: what values of the learning rate provide the fast convergen...

We study the questions concerning the properties and capabilities of computational bithreshold real-weighted neural-like units. We give and justify the two sufficient conditions ensuring the possibility of separation of two sets in n-dimensional vector space by means of one bithreshold neuron. Our approach is based on application of convex and affi...

We introduce the concept of multivalued neural element over Galois field relative to any system of characters of group on which we define logical functions. We deduced the criteria of realization of logical functions on such neural elements. Based on spectral attributes of logical functions the method of synthesis of generalized neural elements ove...

Background. A precise picture of how business processes (in the interpretation by Andrea Burattin) are performed in real-life is vitally important for an organization because it shows actual situation revealing gaps and bottlenecks. Process mining is a discipline with the purpose to research processes using as the input so-called event data (or eve...

Проблематика. У статті розглядається вирішення завдання автоматичного екстрактивного узагальнення тексту на основі рекурентної штучної нейронної мережі з використанням графової інтерпретації тексту й алгоритму оцінки важливості текстової одиниці. Абстрактний підхід набагато складніший, ніж екстрактивний, оскільки він вимагає створення власного вект...

The task of designing information technology for time series forecasting, that bases on fuzzy expert evaluations was considered. A forecasting model, part of which is an expert’s unit, were proposed. The algorithm of synthesis predictive scheme based on the basic predictive models was developed. To determine expert evaluation of the forecast value,...

The paper deals with the problems of Boolean functions realization on neural-like units with complex weight coefficients. The relation between classes of realizable function is considered for half-plane-like activation function. We also introduce the concept of sets separability, corresponding to our notion of neuron. The iterative online learning...

Task of determination structure of social groups was considered as the task of fuzzy clustering. The adaptation of fuzzy c-means was done to solve the problem. Analysis of features design information-analytical system for solving the problem was done. Functional scheme IAS were developed.

This article is dedicated to the development of time series forecasting scheme. It is created based on the forecasting models system that determines the trend of time series and its internal rules. The developed scheme is synthesized with the help of basic forecasting models "competition" on a certain time interval. As a result of this "competition...

r-алгоритму). Ключові слова: оптичне багатошарове покриття, r-алгоритм, матричний метод Абеле, закон Брюстера. The results of theoretical calculations of the optimal four-layer structure for optical bleaching of the substrate when the light is incident at an angle are given in the paper. Computational algorithm is based on the nonsmooth optimizatio...

In this article we have presented development principles of VLSI-structures and the ways to improve effectiveness of VLSI-devices used for coordinated parallel calculation of basic operations of real-time digital signal processing algorithms. We have also developed criteria for the selection of VLSI-structures to be used for such calculations.

The algorithm and specialized VLC-structures for calculus of maximum and minimum values is offered and the estimation of their basic performances is conducted.