Вісник № 02. Системний аналіз, управління та інформаційні технології
Постійне посилання зібрання
Переглянути
Нові надходження
Зараз показуємо 1 - 5 з 16
- ДокументTopic segmentation methods comparison on computer science texts(Національний технічний університет "Харківський політехнічний інститут", 2021) Sokol, Volodymyr Yevhenovych; Krykun, Vitalii Oleksandrovich; Bilova, Mariia Oleksiivna; Perepelytsya, Ivan Dmytrovich; Pustovarov, Volodymyr VolodymyrovichThe demand for the creation of information systems that simplifies and accelerates work has greatly increased in the context of the rapid informatization of society and all its branches. It provokes the emergence of more and more companies involved in the development of software products and information systems in general. In order to ensure the systematization, processing and use of this knowledge, knowledge management systems are used. One of the main tasks of IT companies is continuous training of personnel. This requires export of the content from the company's knowledge management system to the learning management system. The main goal of the research is to choose an algorithm that allows solving the problem of marking up the text of articles close to those used in knowledge management systems of IT companies. To achieve this goal, it is necessary to compare various topic segmentation methods on a dataset with a computer science texts. Inspec is one such dataset used for keyword ext raction and in this research it has been adapted to the structure of the datasets used for the topic segmentation problem. The TextTi ling and TextSeg methods were used for comparison on some well-known data science metrics and specific metrics that relate to the topic segmentation problem. A new generalized metric was also introduced to compare the results for the topic segmentation problem. All software implementations of the algorithms were written in Python programming language and represent a set of interrelated functions. Results were obtained showing the advantages of the Text Seg method in comparison with TextTiling when compared using classical data science metrics and special metrics developed for the topic segmentation task. From all the metrics, including the introduced one it can be concluded that the TextSeg algorithm performs better than the TextTiling algorithm on the adapted Inspec test data set.
- ДокументSoftware testing results analysis for the requirements conformity using neural networks(Національний технічний університет "Харківський політехнічний інститут", 2021) Shepeliev, Oleksandr Vadymovich; Bilova, Mariia OleksiivnaThe relevance of scientific work lies in the need to improve existing software designed to analyze the compliance of the results of software testing of the stated requirements. For the implementation of this goal, neural networks can be used by quality control specialists to m ake decisions about software quality, or project managers as an expert system, for one of the quality indicators for the customer. The article deals with software testing which is a process of validation and verification of compliance of the software application or business program with the technical requirements that guided its design and development, and work as expected, and identifies important errors or deficiencies classified by the severity of the program to be fixed. Existing systems do not provide for or have only partial integration of systems of work with the analysis of requirements, which should ensure the formation of expert assessment and provide an opportunity to justify the quality of the software product. Thus, a data processing model based on a fuzzy neural network was proposed. An approach to allow determining the compliance of the developed software with functional and non-functional requirements was proposed, taking into account how successfully or unsuccessfully implemented this or that requir ement. The ultimate goal of scientific work is the development of algorithmic software analysis of compliance of software testing results to stated requirements for support in the decisions taken. The following tasks are solved in scientific work: analysis of advantages and disadvantages of using existing systems when working with requirements; definition of general structure and classification of testing and requirements; characteristic main features of the use of neural networks; designing architecture, the module of research of conformity of results of testing software to the stated requirements.
- ДокументЗастосування методів штучного інтелекту для апроксимації механічної поведінки гумоподібних матеріалів(Національний технічний університет "Харківський політехнічний інститут", 2021) Погребняк, Сергій Віталійович; Водка, Олексій ОлександровичУ ХХІ сторіччі нейронні мережі широко використовуються в різних сферах, в тому числі в комп’ютерному моделюванні і в механіці. Така популярність через те, що вони дають високу точність, швидко працюють та мають дуже широкий спектр налаштувань. Мета роботи створення програмного продукту з використанням елементів штучного інтелекту, для інтерполяції та апроксимації експериментальних даних. Програмне забезпечення повинно коректно працювати, та давати результати з мінімальною похибкою. Недоліком використання математичних підходів до обчислення та прогнозування петель гістерезису є те шо вони досить погано описують розвантаження, таким чином отримуємо не коректні данні для розрахунків напружено-деформованого стану конструкції. Інструментом вирішення було використання елементів штучного інтелекту, а точніше нейронних мереж прямого поширення. В роботі збудована та навчена нейронна мережа прямого поширення. Вона була навчена вчителем (вчитель з використанням метода зворотного розповсюдження похибки) на основі навчаючої вибірки попередньо проведеного експерименту. Для тестування було побудовано декілька мереж різної структури, які отримували на вхід однаковий набір даних який не використовувався при навчанні, але був відомий з експерименту, таким чином була знайдена похибка мережі за кількістю виділеної енергії та за середньо-квадратичним відхиленням. У статті детально описується математична інтерпретація нейронних мереж, спосіб їх навчання, попередньо проведений експеримент, архітектура мережі та її топологія, метод навчання, підготовки навчаючої вибірки та вибірки тестування. В результаті проведеної роботи було збудоване та протестоване програмне забезпечення в якому використовувалась штучна нейронної мережа, було побудовано та протестоване декілька типів нейронних мереж з різними вхідними даними та внутрішніми структурами, визначені їх похибки, сформовані позитивні та негативні якості мереж які використовувались.
- ДокументEstimating with a given accuracy of the coefficients at nonlinear terms of univariate polynomial regression using a small number of tests in an arbitrary limited active experiment(Національний технічний університет "Харківський політехнічний інститут", 2021) Pavlov, Alexander AnatolievichWe substantiate the structure of the efficient numerical axis segment an active experiment on which allows finding estimates of the coefficients for nonlinear terms of univariate polynomial regression with high accuracy using normalized orthogonal Forsyth polynomials with a sufficiently small number of experiments. For the case when an active experiment can be executed on a numerical axis segment that does not satisfy these conditions, we substantiate the possibility of conducting a virtual active experiment on an efficient interval of the numerical axis. According to the results of the experiment, we find estimates for nonlinear terms of the univariate polynomial regression under research as a solution of a linear equalities system with an upper non-degenerate triangular matrix of constraints. Thus, to solve the problem of estimating the coefficients for nonlinear ter ms of univariate polynomial regression, it is necessary to choose an efficient interval of the numerical axis, set the minimum required number of values of the scalar variable which belong to this segment and guarantee a given value of the variance of estimates for nonlinear terms of univariate polynomial regression using normalized orthogonal polynomials of Forsythe. Next, it is necessary to find with sufficient accuracy all the coefficients of the normalized orthogonal polynomials of Forsythe for the given values of the scalar variable. The resulting set of normalized orthogonal polynomials of Forsythe al-lows us to estimate with a given accuracy the coefficients of nonlinear terms of univariate polynomial regression in an arbitrary limited active experiment: the range of the scalar variable values can be an arbitrary segment of the numerical axis. We propose to find an estimate of the constant and of the coefficient at the linear term of univariate polynomial regression by solving the linear univariate regression problem using ordinary least squares method in active experiment conditions. Author and his students shown in previous publications that the estimation of the coefficients for nonlinear terms of multivariate polynomial regression is reduced to the sequential construction of univariate regressions and the solution of the corresponding systems of linear equalities. Thus, the results of the paper qualitatively increase the efficiency of finding estimates of the coefficients for nonlinear terms of multivariate polynomial regression given by a redundant representation.
- ДокументMetrics of virtual promotion of a product(Національний технічний університет "Харківський політехнічний інститут", 2021) Orekhov, Sergey Valerievich; Malyhon, Hennadiy VasilievichAn approach to the mathematical description of the criterion for the effectiveness of a new object of research – virtual promotion is presented in the paper. The emergence of this new object of research is connected, on the one hand, with the classical theory of marketing, and on the other with modern Internet technologies. Marketing is based on the 4P principle: product, price, location and promotion. Promotion is a component of this principle. But in modern conditions, this phenomenon is changing under the influence of the Internet. Now this 4P component is becoming a fully virtual instrument. The traditional scheme of promotion functioning is as follows. A message is created to a potential buyer and the delivery channel of this message undergoes a change. It is based on the principle: money – goods – money. While the new sales scheme is described by the scheme: we attract a client, make money on a client, we spend money. In the new scheme, we deal with product knowledge in the form of the so-called semantic core of web content. Knowledge describes for a potential client how a given product can cover his need for something. Using the logistic principles of the transfer of goods, this semantic core is loaded into the specified Internet nodes. That is, virtual promotion is formed as two channels: logistics and marketing. The first one performs three operations: concentration, formatting and distribution of semantic cores on the Internet. The second manages this process, forming a virtual promotion map. This map is a graph of Internet nodes. It is required to define such a tree of Internet nodes so that virtual promotion has maximum efficiency. The paper analyzes modern metrics related to the processes of search engine optimization on the Internet. Unfortunately, these metrics evaluate only statistically after the fact of visiting a web resource or the budget of the Internet site in which the advertising message about the product was placed. Therefore, based on the conversion metric, a criterion for the effectiveness of virtual promotion was proposed in the work, which takes into account both the attractiveness of the semantic core and the attractiveness of the Int ernet site where the semantic core will be located. The criterion reflects the income that we receive depending on the attractiveness of the semantic kernel and the Internet site.