ABSTRACT. In today's information age, which is led by technology and science, the essential element in education should be to provide individuals with methods of obtaining information rather than giving information. Self-directed learning is seen as a key competence for survival in the twenty-first century. Self-directed learning is not about learning; instead, it is a meta-theory about learning how to learn. During the pandemic process, many theoretical and applied courses were conducted via distance education. One of these courses is the general chemistry laboratory. While the laboratory course was conducted with distance education, simulation applications were used. For the general chemistry laboratory experiments, first of all, theoretical lectures were made over zoom. Afterwards, the students performed experiments using simulation. The aim of this research is to determine the effect of the general chemistry laboratory conducted with distance education and simulations on the chemistry laboratory self-efficacy perception of pre-service chemistry teachers. The research was designed in a quasi-experimental design model. The data collection tool, which was applied as a pre-test before the research, was made as a post-test after the application. The research was carried out in the spring semester of 2021-2022 in the general chemistry laboratory course. The sample group of the research consists of 25 pre-service chemistry teachers studying at a state university. Data were collected with the chemistry laboratory self-efficacy perception scale. The data obtained from the cchemistry laboratory self-efficacy perception scale were analyzed. As a result of the research, it was determined that the simulation-supported laboratory application had a significant effect on the psychomotor self-efficacy and cognitive self-efficacy of the pre-service teachers. According to the results of the research, pre-service teachers are worried and afraid that they will not be able to do the experiment in the laboratory, so simulations are effective in increasing intrinsic motivation as they are very useful for preparation for the experiment. However, it is an expected result that it is not effective on the development of self-efficacy since it is not a real experimental application. Simulation applications attract attention as effective aids in which individuals will be prepared for the experiment before coming to the laboratory.

ABSTRACT. The article examines the peculiarities of the use of digital tools (DT) in the education of students of the specialty 012 Preschool Education in the discipline «Art needlework». The authors identified the aspects of the study of the problem of using digital tools in the professional training of future teachers, carried out by modern scientists. The relevance of the application of DT for mastering practical disciplines by bachelors in the process of distance learning is determined. It was established: the lack of development of future bachelors in the field of preschool education competence regarding the application of DT for the organization of artistic and productive activities of preschool children. A model of the use of these digital tools in the process of teaching the discipline «Art needlework» has been developed. The experience of using various digital tools of visualization, collective interaction, game services, augmented reality in mastering disciplines is described, such as: Geniallу, Jamboard, Conceptboard, Kahoot, H5P, Сraiyon, Deepdreamgenerator, Dreamstudio, Canva, Fotor, LightShot, Fanny Pho.to, Blippbuilder. The effectiveness of the use of digital tools for mastering practical disciplines in the distance format of the training of bachelors of preschool education has been proven.

ABSTRACT. This paper discussed using a bot-based self-report diagnostic tool to identify post-traumatic stress disorder. The authors present information technology for assessing the psychological state of individuals who have experienced traumatic events in a stressful environment. The technology uses the International Trauma Questionnaire (ITQ) and additional questions describing the current psychological environment. The study employed data analysis techniques to identify significant dependencies between respondents' answers to additional questions about the current environment and their ITQ scores. The bot interface provides a user-friendly platform for respondents to complete the questionnaire. The analytical system, which includes data collection, storage, and processing, allows for flexibility in modifying the questionnaire based on ongoing research.

ABSTRACT.  

The current increase in urbanisation, coupled with the socio-economic motivation for increasing mobility, is pushing the transport infrastructure well beyond its capacity. Traditional urban traffic control techniques are struggling to cope with the dramatic rise of traffic, and have limited ability to react. In response, more intelligent control mechanisms are required to better monitor and exploit the available infrastructure. Despite the growing number of studies leveraging on machine learning techniques to perform traffic control tasks, the good old model-based AI is gaining traction, thanks to its ability to provide approaches that can smoothly deal with unusual and unexpected conditions. In this talk we will focus on the recent application of AI planning to urban traffic control. First, we will look into how urban traffic control is currently performed and what are the main challenges faced. Then, we will present how AI planning techniques have been used to improve common practices, and to provide useful tools for traffic engineers, domain experts, and practitioners.

BIOGRAPHY.

Mauro Vallati is currently Professor of AI at the University of Huddersfield, where he leads the AI4UTMC (AI for Urban Traffic Management and Control https://www.ai4utmc.info/ ) research team. He is an ACM Senior Member, and ACM Distinguished Speaker on artificial intelligence (AI) for the UK. He has extensive experience in real-world applications of AI methods and techniques, spanning from healthcare to train dispatching. Since 2016, he has led several research grants and contracts in the field of urban traffic control, leading to numerous high-impact academic publications, and patents filed in United Kingdom, China, and United States. In 2021 he was awarded a prestigious UKRI Future Leaders Fellowship for investigating AI-based autonomic urban traffic monitoring and control, with the aim of designing intelligent systems that can autonomously recognise the insurgence of traffic congestion and implement traffic light strategies to mitigate its impact on the urban traffic network.

ABSTRACT. The paper is devoted to methods of detecting vulnerabilities of smart contracts using machine learning. The purpose of the study is to improve the accuracy of detecting the reentrancy vulnerability of smart contracts by implementing new machine learning models. A thorough analysis of the current literature was performed and the shortcomings of the existing tools for detecting vulnerabilities of smart contracts were identified. In particular, insufficient accuracy and low adaptability of existing models to new vulnerabilities were noted. To solve these problems, a new model based on a kind of recurrent neural networks with a gating mechanism, namely a bidirectional GRU with an attention mechanism, was proposed to detect the reentrancy vulnerability at the Solidity code level. Using the Word2vec model, the source code of smart contracts was transformed into an array of vectors and used as input to the neural network. Precision, recall and F-beta score were used to evaluate the developed model. 500 smart contract source codes from the Ethereum blockchain were used to train the model, 250 of which had the reentrancy vulnerability. The constructed model was compared with Simple RNN, LSTM, BLSTM, BGRU and BLSTM-ATT models. The obtained results showed that the developed model is ahead of the listed models. The closest values of the metrics were obtained by the BLSTM-ATT model, while the developed BGRU-ATT model uses significantly fewer parameters that need to be optimized, which reduces the training time of the model to detect new vulnerabilities.

ABSTRACT. Machine learning (ML) is opening up new opportunities for the development of innovative systems across a wide range of industries. However, assessing and ensuring the quality of systems with ML components introduces unique challenges related to inherent characteristics of such components like data centricity and unpredictable behavior. Traditional software quality assessment and assurance methods may not be sufficient for ML systems: (1) they focus on software code, while ML systems' quality is influenced by the characteristics of the data and the algorithms used to create ML components; (2) they do not cover the emerging quality characteristics specific to ML systems, such as interpretability, explainability, fairness and trustworthiness. This PhD project aims to develop a comprehensive approach for assessing and assuring the quality of ML systems, with a focus on bias detection and prevention. The research will (1) explore the problem of bias in production ML systems; (2) analyze the gaps in existing software quality models and methods related to bias detection and prevention; and (3) propose an improved approach to quality assessment and assurance to address the challenges associated with bias in ML systems. The results of this PhD project are expected to contribute to the development of better models and methods for assessing and assuring the quality of ML systems, as well as have practical implications for industries that rely on ML systems to automate complex tasks, facilitate decision-making processes and gain insights from large amounts of data.

ABSTRACT. The article describes a bibliographic analysis of available scientific knowledge related to adaptive learning in social sciences. The author selected sources using the Scopus database and conducted a cluster analysis by keyword co-occurrence. The analysis identified five clusters of keywords, including those related to adaptive learning theory, computer-aided instruction, engineering education, didactic fundamentals, and personalized learning. The article also discusses primary concepts, density, and the distribution of concepts over time. Finally, the author limits the analysis years to 2011-2019 and analyze the distribution of concepts during that time.

ABSTRACT. The causal linkage between food security and artificial intelligence has not been fully resolved, whereas AI technologies transformative power has been a key issue in both policy and academic circles for recent decades. At the same time a major shift occurred in understanding the concept of food security from primarily availability, access and utilization to a complex definition adding stability, agency and sustainability parameters. These days the level of food security of states is highly confronted by a number of global threats, namely, conflicts, climate extremes, economic shocks, growing population, pandemics. According to WFP, the scale of the current global food crisis is enormous, with an expected 345.2 million people projected to be food insecure – more than double the number in 2020. Because food crisis is so wide-reaching, there is a strong a need in the transformation of world agriculture and food production sector to alleviate the situation by enhancing food crises management and agricultural productivity through innovative digital technologies, namely AI. Therefore, measuring the impact of AI on food security requires a multi-faceted approach. Drawing on latest research and insights, this study attempted to investigate through data collection, analysis, and country comparison the role that AI can play in addressing food security challenges. The greater the level of implementation of AI in a country, the higher the level of food security of the respective countries from different clusters.

ABSTRACT. The paper is devoted to our research on the effectiveness of using different middle- and long-term trading strategies based on computer analysis (CA) indicators. This paper contains a brief overview of three technical analysis indicators that are usually used for getting buy or sell signals for some specific financial instruments. We have described the approach that allows using such signals not only for a single financial instrument but for a whole investment portfolio. The initial investment portfolio is generated using Harry Markowitz’s Modern Portfolio Theory. The paper contains an overview of similar approaches presented by other researchers. During the experimental part of the research, we compared the effectiveness of using such CA indicators as moving average (MA), relative strength index (RSI), and support and resistance (S&R). The results prove that using certain CA strategies allows not only to increase the initial investment portfolio profitability on rising periods in the financial market but also may reduce loss during a global financial market recession.

ABSTRACT. The reach of online grocery services has expanded to encompass new customer segments in recent years. During the early stages of the COVID-19 outbreak, when delivery slots were limited and customer demand was high, click-and-collect models became increasingly popular. In order to keep pace with evolving customer behavior, it is crucial for retailers to maintain a high degree of operational process efficiency within their business model. This research paper proposes a resource planning system for grocery retail delivery services that utilizes machine learning techniques. The system aims to optimize the allocation of resources, such as delivery drivers, and reduce transport costs, improving the overall efficiency and profitability of the delivery operations. The system is designed to capture and analyze data from various sources, including delivery orders, traffic patterns, weather conditions, and driver schedules. The proposed research demonstrates the potential of machine learning techniques to transform resource planning in grocery retail delivery services and highlights the importance of data-driven decision-making in today's highly competitive retail landscape.

ABSTRACT. Increased access to the stock market leads to a growing interest in investing and an increase in the number of new investors in the market. However, investing in the stock market is not without risks and requires good preparation and analysis. To protect and grow their investments, investors must carefully analyze the company they plan to invest in. Research goal of our paper is to develop a model for predicting financial indicators from annual reports of companies based on data analysis and machine learning using the Python programming language. Tasks of our research contain collection of historical data from annual reports or financial databases; data cleaning, feature selection of relevant explanatory variables (such as revenue, net income, and outstanding shares) to predict the dependent variable (market price per share). The multiple linear regression model showed a strong relationship between the market price per share and the selected explanatory variables with a high coefficient of determination of approximately 0.7225. The coefficients reveal that total equity, net income, and return on equity are positively related to the market price per share, while the current ratio, operating margin, and debt-to-equity ratio are negatively related. However, the magnitudes of these coefficients vary, indicating that some factors may have a stronger influence on stock prices than others may. The conclusion emphasizes the importance of considering additional market factors, such as competition, economic conditions, and industry changes, when making investment decisions. Although the regression model is an effective tool for analyzing the dependence of stock prices on various variables, it may not account for all factors affecting stock prices. Therefore, investors should use market analysis as a whole to make the right investment decisions.

ABSTRACT. In this study, we apply the multifractal detrended fluctuation analysis concepts to financial time series that helps to determine the onset of a crash in the Dow Jones Industrial Average index. For the studied index we emphasize 4 the most influential stock market crashes: Wall Street Crash of 1929, Black Monday of 1987, Financial crisis of 2007–2008, and 2020 stock market crash. We present that economic crashes on a mapping with multifractality phenomena demonstrate a dynamic phase transition. Some of the presented multifractal measures appear to be analogues of free energy and specific heat, and can be used as indicators or indicators-precursors of the stock market crashes.

ABSTRACT. The development of cognitive technologies and the formation of competences become an integral part of success in the conditions of a rapidly changing world. This document presents a bibliometric analysis of cognitive technologies and competencies. The dataset was retrieved from the Scopus database, analyzed and presented using VOSviewer. The search equation identified 281 studies. After analysis and application of filters, 60 studies were selected for further analysis. Among the journals that publish research on this topic, the most productive is Ceur Workshop Proceedings (5%). The most cited authors are González-González and Jiménez-Zarco. Authors from Indonesia, Spain and China published the most articles. The top 5 thematic categories include Computer Science; Social Sciences; Engineering; Business, Management and Accounting; Energy. The most cited article is Birjali, M., Beni-Hssane, A., Erritali, M. Research on cognitive technologies and the development of competencies open broad perspectives for improving learning, professional development, and personal growth.

ABSTRACT. The article “Smart-Systems in STEM Education” explores the significance of integrating "Smart-systems" technologies into STEM (Science, Technology, Engineering, and Mathematics) education. The article highlights the role of "Smart-systems" in preparing students for the future by providing practical experiences and fostering critical thinking and problem-solving skills. It discusses various technologies associated with "Smart-systems," such as robotics, IoT, AI, and data analytics, and their applications in STEM education. The article also presents a range of resources available for educators and students, including online courses, educational websites, maker spaces, and competitions. By leveraging these resources, educators can create engaging learning environments that inspire students to explore and pursue careers in emerging fields. The article emphasizes the benefits of incorporating "Smart-systems" in STEM education, including the development of technological literacy, interdisciplinary learning, and the cultivation of skills necessary for the digital age. Ultimately, embracing "Smart-systems" in STEM education empowers students to become the next generation of innovators and problem solvers who can contribute to a rapidly evolving technological landscape.

ABSTRACT. The article considers the possibilities of using cloud technologies to create an effective educational environment is aimed at the development of future primary school teachers' soft skills. The experience of using cloud technologies is demonstrated, the advantages, challenges and possible limitations of their use are revealed. Prospects for further development of the topic are determined. The main aspects of the article include: clarification of the essence of the concept of “educational environment” and signs of a quality educational environment; review of the potential of cloud technologies for the development of future primary school teachers' soft skills; characteristics of the SAMR model, which describes the levels of technology integration in the educational process; systematized data on types of activities, their impact on the development of certain Soft Skills of future primary school teachers and applied cloud technologies with practical implementation examples. The obtained research results are emphasized the potential of cloud technologies in creating an effective educational environment is aimed at the development of soft skills of future primary school teachers. The conclusions of the article can be served as a basis for further research and the development of programs and strategies aimed at the effective use of cloud technologies in the educational process.