Headings of the journal
"Educational Resources and Technologies"

Annotation: The article describes the problem of optimizing the execution plans of analytical SQL queries using machine learning. Currently, none of the popular database management systems (DBMS) uses an optimizer based on machine learning. At the same time, it is machine learning that can solve the known problems of standard optimizers used in most modern DBMS. In this work, the disadvantages of existing approaches to planning analytical SQL queries are considered in detail. There is a lack of research on the use of well-known machine learning-based approaches to optimizing SQL queries in analytical databases. The relevance of solving the problem of optimizing SQL query execution plans for massively parallel column-based DBMS is shown. The paper provides a justification for the choice of specific methods based on machine learning. The modifications made are described, in particular, an approach to the selection of execution plans for analytical SQL queries based on the collection of a training sample and the selection of target values is proposed. The problem of implementing a machine learning-based approach is considered. Possible problems with the training of the model and its operation in fundamentally new situations arising during training are highlighted. The ways of their solution are proposed. Original experiments have been carried out with various methods of optimizing SQL queries in relation to a massively parallel column DBMS. It is shown that the proposed modifications of existing solutions significantly improve the speed of query execution.

Annotation: This study considers the metaphors used in the Russian and Chinese worldviews to characterize the teaching profession. The purpose of the work is the presentation of the diachronic and synchronic analysis of the metaphorical portrait of the teacher, starting with the works of Ya.A. Komenskiy to the present day, as well as the results of an empirical study of the reflective potential of the portrait of a teacher in the process of educating foreign language teachers. The article analyzes the views not only on the positive aspects of the teacher’s activity, but also critically examines his/her possible negative impact on the development of the personality of students. In the empirical part, the emphasis is on the results of a survey of foreign language teachers who have taken refresher courses in this area. The results of the study demonstrate the versatility of the concept of “teacher” in the Russian and Chinese worldviews, as well as the variability of respondents’ opinions regarding the role of the teacher in the modern model of teacher training. The results obtained can be interpreted for further projecting of a program for the professional development of foreign language teachers.

Annotation: Currently, transformer-based transfer learning methods are used to solve natural language processing problems, but they can be demanding on computing resources and memory. An alternative approach proposed in this article is related to the use of a pre-trained language model based on the recurrent neural network LSTM (Long Short- Term Memory), which allows natural language processing (tonality analysis) to be performed on a set containing text data. As a reference test of the proposed model for solving the problem of recognizing the “hate speech” in the text, other models based on transformers are considered. The degradation test, which evaluates the resilience of models to performance degradation with a decrease in the amount of training data, confirms the effectiveness of the model for a limited training dataset. The proposed LSTM model based on a recurrent neural network allows us to use a pre-trained language model from scratch and provide faster pre-learning in transformer-based models.

Annotation: The article describes the comparative effectiveness of automated credit risk assessment by modern machine learning methods (Naive Bayes classifier, k-nearest neighbors, logistic regression, “random forest” and deep neural network). A database was modeled from available open sources, samples were made, and data was preprocessed. Machine learning models have been trained and tested, and compared by quality metrics. Based on the test results, the parameters of the machine learning model were optimized by a complete search of combinations of statistical models (rule-based method, k-nearest neighbor method, logistic regression, discriminant analysis, naive Bayes classifier, neural networks and decision trees). Each of the traditional classifiers (naive Bayes classifier, k-nearest neighbor method, logistic regression, “random forest”, deep neural network) was improved with the use of standardization mechanisms and dimensionality reduction by the principal component method. Based on the results of a comparative analysis of machine learning models, the best result was demonstrated by a neural network model with optimized parameters, which provided the best indicators for all evaluation metrics.

Annotation: The article examines the problem of information security of smart systems. The level of digitalization of smart systems is continuously increasing and vulnerability factors and threats are changing accordingly. Therefore, the information protection process needs to be continuously improved. The purpose of the work is to systematize the threats and vulnerabilities of modern systems of smart things, to develop recommendations for improving their information security. The systematics of threats to Wi-Fi and Bluetooth wireless technologies is presented. An example of vulnerabilities of a smart system when using Wi-Fi technology is considered. The typological series of Wi-Fi technology is presented, which is created according to the degree of the transmitted information security growth. The threats and vulnerabilities of the technology when using specific Wi-Fi network protection protocols are shown. The example shows that information threats appear not only because of the vulnerabilities of wireless information transmission technology, but also because of the technological features of the use of this technology in other systems, including “smart things”. A comparison is made of the security of Bluetooth and Wi-Fi technologies based on the use of the modern WPA3 protocol. Recommendations have been developed for the transmission of confidential information via Wi-Fi and Bluetooth channels.