Based on the synthesis of complex analysis methods, perturbation theory, and characteristics, a new approach has been developed for accounting for osmosis and temperature in predicting the migration processes of radionuclides under non-isothermal conditions in quasi-ideal porous media (in curved regions) bounded by streamlines and equipotential lines. The solution to the corresponding degenerate problem was obtained based on the developed method of nonlinear inversion of boundary value problem solutions into quasi-conformal mappings. An algorithm for calculating a uniformly dynamic grid, streamlines, and speed of ideal filtration field was constructed. With the developed method of characteristics, formulas for the approximate solution of diffusion-convection mass transfer problems in perturbed osmosis and heat filtration fields were obtained.
This article aims to discover how AI-powered systems facilitate auditing, what risks emerge for AI-assisted audits and how to deal with these new risks. The paper studies the impact of cognitive computing on audit risk. AI-powered software is capable of self-learn so that it can identify patterns in data and codify them in predictions, rules and decisions. This self-learning ability can become both a benefit and, at the same time, insecurity. Although AI-self-learning helps make the process more efficient and calculations more accurate by improving the algorithm, eliminating errors and reducing risks, it creates new previously unknown threats. We discovered inherent limitations of cognitive-based technologies and risks for the audit process associated with using AI systems. We also proposed a complex security model that can reduce the uncertainty of AI-enabled audit and provides insight into future research opportunities.
This research demonstrates the effectiveness of using project-based learning to develop virtual and augmented reality applications using available platforms such as Vuforia and Unity 3D. In project-based learning, students acquire knowledge and skills by researching and responding to interesting and complicated questions, problems or issues. The research involved 54 students from two universities. The students who participated in the research were separated in an experimental and a control group. Control work of the same level was carried out in both groups. In the post-test, 56% of the participants in the experimental group scored above 93, while in the control group only 36% of all participants scored above 80. The research showed that the knowledge and skills acquired by students in the experimental group in developing applications of virtual and augmented reality were higher than in the control group.
This paper is aimed to the analysis of the concept of self-improvable computer systems design through the use of reconfigurable hardware, specialized processors high-level automatic design and synthesis tools, and Artificial Intelligence technologies. It highlights FPGA as the hardware basis for self-improvable computer systems and considers the development directions for FPGA-based computer systems in the future. The Artificial Intelligence technologies are used to identify what and how the computer system can be improved by itself. The software for specialized processors high-level design tools serves as the basic acting mechanism for computer system self-improvement. Finally, the method of computer system self-improvement is presented in the paper.
Control of users’ access to the information system is the main part of the security system. Moreover, it must work in real time and be easily reconfigured. Such controller can be built by using fuzzy logic. In this paper, the authors propose a fuzzy controller for access distribution to the student assessment system. This approach allows applying the proposed fuzzy controller to any evaluation system, as it can be easily reconfigured. Hardware implementation of this controller allows to increase the stability of the information security system because of subjectivity exclusion.
The process of unsupervised generative learning with visual data and the structure of informative low-dimensional sparse generative representations of images of handwritten digits were investigated. Learning models with the architecture of a sparse convolutional autoencoder with constraints to produce low-dimensional representations achieved successful learning demonstrated by training metrics and high accuracy of generation of images of digits. A well-defined, continuous and connected “stacked” structure of low-dimensional slices in the sparse latent space produced by activations of participating latent neurons was observed and described in detail. The conclusion is that structured informative representations obtained with unsupervised generative models can be an effective platform for investigation of the emergence of common types or “concepts” in sensory inputs in artificial and biological learning systems.
Power-to-gas (P2G) is a key technology for the energy transition. However, data on biogenic resources and surplus renewable electricity needed for sector coupling is either not available at all, incomplete, barely differentiated and not standardized. The project DanuP-2-Gas tackled this issue by analysing the data availability and needs for the Danube Basin, collecting data and subsequently visualizing it in an GIS-based Atlas Tool. Main challenges were the lack of uniform data and the lack of spatial differentiation of the biomass availability. Still, an overview of the general feasibility of P2G in the Danube region countries could be provided.
Reducing greenhouse gas emissions into the atmosphere requires significant expenditures, the increase of which will reduce the negative impact on the environment and positively affect the innovative-social component of the development of the economy and society. This article analyzes the impact of expenditures on reducing greenhouse gas emissions using panel data from EU countries. A step-by-step multifactor analysis was applied to identify the most significant factors. Canonical correlation analysis was used to identify the simultaneous influence of national expenditures and investments in environmental protection on the volumes of greenhouse gas and carbon dioxide emissions.
The goal of this paper is to enhance text summarization using a hybrid methodology. The process of producing a condensed version of a text while keeping its essential details is known as text summarization. In this study, we have presented a method for training the T5 model on the SAMSum dataset with conversation sentences to increase its effectiveness in text summarization. To determine the impact of training the T5 model on the dataset, the model is assessed using the ROUGE metric both before and after training. ROUGE is a set of metrics used to evaluate the quality of automatic summaries by comparing them to reference summaries based on the overlap of n-grams, word sequences, and other linguistic units. In order to enhance the quality of the generated summary, our hybrid approach makes use of extractive and abstractive summarization techniques as well as pre-and post-processing techniques. There is an improvement in the ROUGE metrics of the model before and after training. ROUGE1 before training was observed to be 25.53 while ROUGE1 after training is calculated to be 45.17.
In this study, we investigated the ethical principles of trustworthy AI and differentiated five prime factors essential for developing trust in AI and most widely presented in regulatory guidelines worldwide. By utilizing Fuzzy Logic Toolbox in MATLAB 9.4, we evaluated the impact of primary ethical principles on trustworthy AI systems in a systematic and structured manner. We discovered that the principle of Fairness and Non-discrimination is the most influential for the development of trustworthy AI, as it is the most represented in the regulatory guidelines. The proposed model offers two main benefits for developers and deployers of AI systems, including predicting the potential public trust in AI systems and assessment compliance with the regulatory frameworks. To ensure the continued trustworthiness of AI systems, the model should be used at all stages of the software life circle, including during development, before placing the system on the market, and at the stage of use to monitor compliance with the safeguards declared to users.
This paper deal with software development for decision-making support of the Unmanned Aircraft Systems (UAS) operators when preparing to flight as well as for weather hazards avoidance during the UAS flight and mission fulfillment. The developed software also can be used for meteorological data obtaining, collection, processing, exchange and dissemination. The general software architecture for decision-making support is proposed and analyzed. The interface of the software is demonstrated. The abilities of the developed software are discussed. The decision-making in the proposed software is made on the base of a risk-oriented approach.
The article discusses the approach to the identification of the recurrent laryngeal nerve (RLN) during surgery on the thyroid gland. This technology consists of the following main stages: obtaining an information signal as a result of stimulation of the tissues of the surgical wound with electric current; signal segmentation - response to irritation; selection of the main spectral component of segments of the information signal; modeling of the distribution of the amplitudes of the main spectral components for the identification of RLN based on the use of an interval discrete model in the form of a difference scheme. The most problematic is the last stage, which is based on the construction of a mathematical model that reflects the electrophysiological properties of the area of the patient’s surgical wound. To configure this model, it is proposed to use an ontological approach. The ontological description of thyroid surgery makes it possible to set up a differential scheme for the identification of RLN taking into account the specifics of the patient’s disease and thus shortens the time for examining the surgical site.
The article presents an approach to automating the processes of controlling the parameters of floating docks at the local and remote level. The hierarchical functional structure and hardware and software of the remote parametric control system of a floating dock are described. The presented computer system performs decentralized processing of information between the SCADA system and cloud services.
The effectiveness of the developed structural and functional model of forming intercultural competence in future specialists in the field of “Information Technology” in the educational environment of the university was experimentally tested and described in the article. The implementation of the structural and functional model was carried out in three stages: orientation and value (awareness of the importance of knowledge and skills of intercultural competence and its formation as a personal value by future specialists in the field of “Information Technology”; formation of values of intercultural competence); educational and cognitive (mastering knowledge of intercultural competence in order to use it in the process of future professional activity; familiarization with the specifics of professional training of future specialists in the field of “Information Technology”). A comparative analysis of the results of the formative and control stages of the experiment showed a significant increase in the levels of intercultural competence of future specialists in the field of “Information Technology” of students of the experimental group, which is evidence of sufficient effectiveness of the developed structural and functional model of formation of intercultural competence of future specialists in the field of “Information Technology” in the educational environment of the university and the achievement of the goal at the beginning of the study. The analysis of domestic and foreign scientific literature has shown that researchers have considered problems related to the formation of intercultural competence in future specialists in the field of “Information Technology in particular, they have studied: culture as a specific phenomenon, the use of a competence-based approach in education, the essential characteristics of the phenomenon of intercultural competence, the peculiarities of the formation of intercultural competence of the individual, the functioning of the educational environment of a higher education institution.
The paper shows the possibility of using texts related to hashtags for the automated classification of texts from business pages or organizations on the social network Instagram. Three groups of texts for neural network training related to the hashtags “IT”, “medicine” and “astronomy” have been identified on Instagram. Texts from Instagram posts of “microsoft”, “med.base” and “nasa” accounts are collected as a test data set, respectively. The best results of post-categorization are observed for the field of information technology, fewer texts are specified by words in the medical field with hashtags as trained data. The classifier can be used to automatically sort content or determine the probable belonging of pages to a defined set of categories.
Objectives: The important task of analog-to-digital conversion in distributed management information systems has been defined. The need for the application of new conversion methods to increase the speed of converters has been substantiated. Methods: Known methods of additive successive approximation have been analyzed, and a new method of bitwise balancing has been proposed. A mathematical model has been constructed, and algorithms for analog-to-digital conversion have been investigated. Results: A comparison of the known and proposed methods has been made, and it has been determined that the proposed method allows for a reduction in conversion time by 6 to 25%. Conclusion: The improved indicators of analog-to-digital conversion have determined the perspective of technical means development and application for accelerated information form conversion.
This paper is devoted to developing of a load balancing model between several databases with potential for their unlimited scaling.
The article explores a significant scientific challenge related to the development of techniques and tools for constructing discrete models of complex objects using interval difference equations. This approach combines ontological principles with the analysis of interval data to broaden the range of applications for these models. It serves as a catalyst for advancing applied research in fields such as national defense, environmental protection, medicine, and other domains where mathematical models of objects with distributed parameters play a crucial role in decision support systems. The fundamental aspect of this mathematical modeling approach, based on interval analysis, involves multiple parameter estimations for input-output models. These models are constructed using experimental data, where output variables are expressed in interval form. The key research findings outlined in the article include: An explanation of how the ontological approach to mathematical modeling, using interval data, can be applied for software development. This application aims to expand the range of scenarios in which models can be used while maintaining their predictive capabilities. A step-by-step framework is proposed for creating an ontology-driven software system for mathematical modeling based on interval analysis. This framework outlines the process of developing, utilizing, and updating the ontological model of the subject area of mathematical modeling with interval data. One distinctive feature of the approaches discussed in this article is their adaptability as software extensions to applied systems of mathematical modeling that utilize interval analysis. By combining interval analysis and ontological representation of the subject area, these approaches enhance the efficiency of computational procedures for identifying models of complex objects. They also allow for the adaptive utilization of different models in various subject areas within decision support systems.
The study aims to substantiate the significance of future social workers’ ICT competence formation and identify their influence on the professional activity and effectiveness of their professional training and work. The participation of future social workers in social work with various social work clients using ICT is multifunctional. It is versatility, knowledge of the regulatory and legal framework, and the presence of appropriate personal qualities for establishing trusting relationships with various categories of social work clients. The professional activity of future social workers is determined by a set of unique attributes, value orientations, and interests, which significantly influence the system of professional interaction with various social work clients. It is highlighted that the traditional organization of professional training of future social workers in institutions of higher education isn’t fully provided without opportunities of using digital technologies for the formation of information competence of future social workers. The effectiveness of the proposed pedagogical conditions is confirmed by introducing several diagnostic methods, which made it possible to reveal how effective the developed pedagogical conditions are to carry out their further implementation. It is noted that the results of the experiment proved their suitability for the implementation of ICT competence in the educational process of higher education institutions for the training of future social workers.
The article considers the practical experience of introducing interactive adaptive learning in the e-learning environment. The basis of the learning environment is the LCM Moodle learning resource management system. One of the ways to individualize and personalize the learning process is to attract interactive and adaptive technologies for teaching students. The LCM Moodle’s interactive module of activity “Lesson’’ is a series of HTML pages linked by transitions. The main difference between “Lesson’’ and other LCM Moodle modules is that there are elements of adaptability. Using this tool, each student’s choice can be accompanied by appropriate comments from the teacher and the ability to go to different pages of the course depending on the correctness/incorrectness and completeness of the answers. With such planning, “Lesson’’ can provide theoretical material and control tasks to check its mastery for each student automatically, without additional action by the teacher. As a result of completing the tasks, the corresponding score appears in the student’s register of marks. In general, this module of activity is similar to a classroom lesson, when the teacher, presenting new material, from time to time interviews students in order to identify the level of its mastery.
