To determine the most effective approach for solving the task of classifying the emotional tone of a message, we trained selected neural network models on various sets of training data. Next, based on the performance metric of the percentage of correctly classified responses on a test data set, we compared combinations of training data sets and various models trained on them. During the writing of this article, we trained four neural network models on three different sets of training data. By comparing the accuracy of the responses from each model trained on different training data sets, conclusions were drawn regarding the neural network model best suited for solving the task at hand.

n the modern world, streaming data has become widespread in many subject areas. The task of processing streaming data in real time, with minimal delay, is highly relevant.

In stream processing, data processing, various approximate algorithms are often used, which have much higher time and memory efficiency than exact algorithms. In addition, there is often a need to forecast the state of the stream.

Thus, there is currently a need for a tool for sequential snapshotting of aggregated data from streaming data, enabling flow state prediction and approximate algorithms for stream data processing.

The authors of the article have developed such a tool, reviewed its architecture and mechanism of functioning, and evaluated the prospects for its further development.

The effectiveness of the ImageNet diffusion model and CLIP models for image generation based on textual descriptions was investigated. Two experiments were conducted using various textual inputs and different parameters to determine the optimal settings for generating images from text descriptions. The results showed that while ImageNet performed well in generating images, CLIP demonstrated better alignment between textual prompts and relevant images. The obtained results highlight the high potential of combining these mentioned models for creating high-quality and contextually relevant images based on textual descriptions.

This article presents a new algorithm for 3D reconstruction using a combination of two existing methods – Structure from Motion (SfM) and Oriented FAST and Rotated BRIEF (ORB). The authors propose an approach that merges the advantages of both methods to enhance the accuracy and efficiency of reconstructing the 3D structure of scenes from images. To improve reconstruction quality, filtering and outlier removal are applied, along with other optimizations. Comparative results between the new algorithm and existing methods demonstrate its superiority in accuracy and noise robustness. The proposed approach is highly scalable and can be successfully applied in various fields that require precise 3D reconstruction of image scenes.

A novel approach to supervised face generation using open-source generative models including StyleGAN2 and Ridge Regression is presented. A methodology that extends StyleGAN2 to control facial characteristics such as age, race, gender, facial expression, and hair attributes is developed, and an extensive dataset of human faces with attribute annotations is utilized. The faces were encoded in 256-dimensional latent space using the StyleGAN2 encoder, resulting in a set of characteristic latent codes. We applied the t-SNE algorithm to cluster these feature-based codes, demonstrated the ability to control face generation, and subsequently trained Ridge regression models for each dimension of the latent codes using the labeled features. When decoded using StyleGAN2, the resulting codes successfully reconstructed face images while maintaining the association with the input features. The developed approach provides an easy and efficient way to supervised face generation using existing generative models such as StyleGAN2, and opens up new possibilities for different application areas.

The article is devoted to the development of a search and indexing system for audio files using Automatic Speech Recognition (ASR) and Elasticsearch. Current Russian-language audio file transcription systems have been analyzed, and Whisper has been chosen as the best one. An algorithm for optimizing transcription speed using parallelization of file processing processes has been developed, and its effectiveness has been demonstrated. A microservice architecture-based system has been built, capable of indexing audio file content and their metadata for search purposes. The research results show that the proposed approach can be applied to create efficient and flexible systems for searching and analyzing audio information.

In this article, we have described expirience of the new NoSQL Database Management System (DBMS) development named as “GoldenRaceDB”. Also, we have described prerequisites that needed to create it in the context of the Russian software’s import problem. The new technology is realised as based on server environment Node.js. To understand this article, a reader should have expirience of server-side development by using one of high-level progrmming language (as minimum) or expirience of development custom DBMS. This technical solution is not open source, we have created it to solve our local tasks at our organization exactly, the technology birth place. But a reader can understand a general vector of creating custom resized DBMS by using other high-level programming language.

The article describes the role of museum data in scientific geological research, as well as the integration of this data into the digital space of geological knowledge for more efficient use and analysis of distributed geological and museum resources and the possibility of building digital models.

We present a set of semantic services developed by us to support the educational process in mathematics. The functionality of these services is based on the use of mathematical ontologies OntoMath^Edu and OntoMath^PRO. The ontology of professional mathematical knowledge OntoMath^PRO is designed to classify and systematize the concepts of professional mathematics and includes several important areas of mathematics. Educational mathematical ontology OntoMath^Edu systematically represents knowledge on the training course “Planimetry”. For the use of ontologies in educational applications, an approach to the design of prerequisite relations in these ontologies has been developed. To support mathematical education, we have developed: a service for semantic search by mathematical formulas, a service for semantic annotation of educational materials, a service for visualizing subgraphs of the OntoMath^Edu ontology semantic network, a parallel formal/informal corpus of mathematical statements, a system for automatically generating test questions in mathematical disciplines.

We provide examples of successful application of the developed software tools.

The created software tools are built into the OntoMath digital ecosystem. This ecosystem implements the interaction of semantic services for managing mathematical knowledge.

The incredible growth in the popularity of digital technologies and education has led to the emergence of “computational thinking”. The dramatic demand and popularization of this notion ignited the redesign of educational standards and the upsurge of literature on this subject. During the extensive literature review, we found that it is not always clear what computational thinking really is, what aspects it entails, and how to think about it in simple yet robust terms. In this paper, we have examined the meaning of this concept, its importance in modern digital education, and drawn an analogy between computational thinking and the skills of writing ordinary text to facilitate its reliable understanding.