Topic profiles represent publications, authors, and other scholarly entities as probability distributions over a fixed set of topics, but flat total variation treats every pair of distinct pure-topic profiles as maximally separated and therefore ignores taxonomic proximity. From a rooted weighted taxonomy, we derive a cardinality-normalized linear operator that maps the leaf topics to points in the original probability simplex and exactly realizes a normalized lowest-common-ancestor ultrametric under total variation. The operator is doubly stochastic and positive definite; within the class of nonnegative edge-cluster Gram operators, its normalization is uniquely determined on the reduced branching tree. Applying the same invertible operator to arbitrary topic mixtures yields a nondegenerate hierarchy-aware metric that contracts flat total variation, differs from the tree-Wasserstein distance on mixtures, and can be evaluated in O(|V|+L) time and memory without forming the dense matrix. In a frozen OpenAlex taxonomy with 4,516 terminal Topics, raw dissimilarities between Topic texts showed consistent ordinal alignment with taxonomic proximity, while only 3 of 253 calibrated internal nodes required monotonic correction. Encoder choice nevertheless affected individual height estimates. The framework exactly realizes a supplied weighted hierarchy; text is used only to initialize its node heights, and distances between scholarly topic profiles are then computed in the induced geometry.

This article describes methods for analyzing scientific activity and the tools developed on their basis to support informed decision-making at all levels of management in the scientific domain — from individual researchers to the heads of research organizations. A review of domestic and international studies in the field of scientometrics is provided, highlighting key trends and identifying existing gaps that the Institute of Control Sciences of the Russian Academy of Sciences (ICS RAS) project seeks to address through the creation of an information system for the analysis of scientific activity. The project involves the development of models, methods, and algorithms for analyzing the scientific activity of researchers, organizations, journals, and conferences in the field of control theory, using publication data. This problem is approached through an interdisciplinary framework that integrates network analysis, ontological design, big data processing, machine learning techniques, and optimization methods. The proposed system is built upon an ontology of scientific knowledge in control theory, the main principles of its construction and structure being discussed in this article. Combined with the system's technological capabilities, the presented ontology enables a level of analytical detail that surpasses existing scientometric systems (such as Web of Science, Scopus, Russian Science Citation Index, etc). It is demonstrated that the principles underlying the development and structure of the proposed ontology can be applied to other research domains. The paper also considers modern approaches to data collection and visualization, as well as methods of network and content analysis of scientific texts.The main concepts, methods, and models for the analysis of scientific activity are summarized, and their implementation within the developed information system is discussed.

Centrality metrics play a crucial role in network analysis, while the choice of specific measures significantly influences the accuracy of conclusions as each measure represents a unique concept of node importance. Among over 400 proposed indices, selecting the most suitable ones for specific applications remains a challenge. Existing approaches -- model-based, data-driven, and axiomatic -- have limitations, requiring association with models, training datasets, or restrictive axioms for each specific application. To address this, we introduce the culling method, which relies on the expert concept of centrality behavior on simple graphs. The culling method involves forming a set of candidate measures, generating a list of as small graphs as possible needed to distinguish the measures from each other, constructing a decision-tree survey, and identifying the measure consistent with the expert's concept. We apply this approach to a diverse set of 40 centralities, including novel kernel-based indices, and combine it with the axiomatic approach. Remarkably, only 13 small 1-trees are sufficient to separate all 40 measures, even for pairs of closely related ones. By adopting simple ordinal axioms like Self-consistency or Bridge axiom, the set of measures can be drastically reduced making the culling survey short. Applying the culling method provides insightful findings on some centrality indices, such as PageRank, Bridging, and dissimilarity-based Eigencentrality measures, among others. The proposed approach offers a cost-effective solution in terms of labor and time, complementing existing methods for measure selection, and providing deeper insights into the underlying mechanisms of centrality measures.

This paper presents a comprehensive analytical review of contemporary mathematical models of information influence and control in social networks, emphasizing the integration of agent-level factors such as trust, reputation, decision-making, and action execution. We introduce extensions to classical models, including the DeGroot model, by incorporating these critical components to more accurately reflect social interactions. Special attention is given to control strategies and the application of game-theoretic methods for analyzing interactions among controlling entities. We apply these models and methods to real-world social network data, including analyses of ideological preferences and public opinions on health measures during the COVID-19 pandemic. Our approach provides new perspectives for future research in social process modeling and offers insights for designing effective strategies for information influence and control.

Представлено описание подходов, лежащих в основе разрабатываемой в ИПУ РАН Информационной системы анализа научной деятельности (ИСАНД) в области теории управления. Описана онтология ИСАНД, ориентированная на представление и сбор знаний в области теории управления: как научного знания (онтология теории управления), так и знаний, связанных с научной деятельностью агентов в данной области (организаций, журналов, конференций и отдельных исследователей). Дана схема построенной на основе онтологии архитектуры ИСАНД как сложного программного комплекса, обеспечивающего сбор, хранение и анализ публикаций и их метаинформации, которые поступают из внешних источников. Описан алгоритм построения тематических профилей научных объектов (публикаций, ученых, организаций, журналов, конференций), описаны осуществляемые при помощи ИСАНД процессы обработки текстов и возможности сетевого анализа. Описаны основные возможности использования ИСАНД. This paper describes the approaches underlying ISAND, an information system for scientific activity analysis in the field of control theory and its applications. ISAND is being developed at the Trapeznikov Institute of Control Sciences, the Russian Academy of Sciences. The ISAND ontology is oriented toward the representation and collection of knowledge in the field of control theory and its applications, namely, scientific knowledge (the ontology of control theory) and knowledge related to the scientific activity of agents (organizations, journals, conferences, and individual researchers) in this field. Based on this ontology, the ISAND architecture is a complex program system to collect, store, and analyze publications and their metadata from external sources. The ISAND algorithm for building the thematic profiles of scientific objects (publications, researchers, organizations, journals, and conferences), as well as ISAND text processing and network analysis capabilities, are presented. Finally, the main possibilities of using ISAND are considered.

This paper examines a simulation model of information cascade formation (a sequence of comments) on a social network post. It assumes that a user has two parameters characterizing their opinion and action. User opinions are influenced by previously written comments (including those by bots) that the user sees. Additionally, some comments may be removed during moderation.

Examining the possibilities of mutual influence of users in new media, the researchers found a high level of aggression and hate speech when discussing an urgent social problem - measures to fight the COVID-19 pandemic. This fact determined both the central topic of the proposed article and the pivotal aspect of the follow-up research. The first chapter of the article is devoted to the characteristics of the prerequisites of the undertaken research and its main features. The subsequent chapters include methodological hallmarks of the study, theoretical substantiation of the concepts of "aggression" and "hate speech" and identification of systemic connections of these concepts with other characteristics of messages. As a consequence, a mathematical model of aggression growth was created and its manageability was analyzed using basic social media strategies. The results can be useful for developing media content in a modern digital environment.

We examine a model that focuses on forming an information cascade of user comments on a post within an online social network. These comments exhibit various attitudes toward a specific issue and can be categorized as positive, negative, or neutral. The probability of a user composing a comment with a particular attitude is influenced by their initial opinion and the collection of previously posted comments they encounter. The arrangement of these comments in a specific order is determined by the social network's algorithm. Through simulation modeling, this study investigates the impact of the algorithm on the properties of the resulting information cascade.

A number of methods and examples are considered that substantiate the possibility andexpediency of using the actional approach to solving applied problems of analyzing informationprocesses in active network structures.

The model of mixed types of agents in social networks is considered. For this model, the problems of information control and confrontation are solved. Two cases of confrontation are simulated: with the information interaction of network agents (the joint dynamics of their opinions and actions) and without information interaction. Equilibria are calculated for these situations.

Analyzing the possibilities of mutual influence of users in new media, the researchers found a high level of aggression and hate speech when discussing an urgent social problem - measures for COVID-19 fighting. This fact determined the central aspect of the research at the next stage and the central topic of the proposed article. The first chapter of the article is devoted to the characteristics of the prerequisites of the undertaken research, its main features. The following chapters include methodological features of the study, theoretical substantiation of the concepts of aggression and hate speech, identification of systemic connections of these concepts with other characteristics of messages. The result was the creating of a mathematical aggression growth model and the analysis of its manageability using basic social media strategies. The results can be useful for developing media content in a modern digital environment.

A strong acoustic shock as a result of the outflow of a supersonic jet when discharging technical gases causes serious damage to ground facilities and the health of personnel. This leads to a significant decrease in the safety level of technical operation facilities. Therefore, the problem of reducing aerodynamic noise during supersonic jet outflow is urgent. The reducing possibility the aerodynamic noise of a supersonic jet is considered experimentally. The jet outflow from a long channel in which a pseudoshock is formed was considered. Two types nozzles were located at the end of the outflow channel. The first type nozzles with longitudinal slots and the second type are multichannel (multitube). In the experiment, the total noise level from the outflowing supersonic jet and the change in static pressure along the length of the outflow channel were measured.

Detailed data on the measurement of the gas-dynamic parameters of the flow in the jet field at a large value of the relative total pressure (Npr = 70) are presented. A significant increase in the measured total pressure in the first cell of a supersonic weakly underexpanded jet has been experimentally revealed. An explanation is given of this effect associated with the mixing process in the region behind the Mach disk of the peripheral high-pressure flow and the axial flow behind the Mach disk. This explanation based on the data of the numerical calculation of the flow. Experimental data can be used to verify the results of numerical calculations.

We consider a multidimensional model of opinion dynamics in social networks. Within theframework of the model, the dynamics of two interconnected information processes in a socialnetwork is studied. The first process is the process of spreading of the excitation in the network ofagents and their actions observed from outside (e.g., in the form of messages posted in socialmedia). The second process, which has a connection with the first, is the formation of agents’opinions (which are a characteristic of their internal state). We demonstrate that the proposedmodel of opinion dynamics is flexible and allows taking into account the significant effects ofopinion formation in social networks, including consensus or agreement of opinions, preservation ofdifferences in agents’ opinions, and even polarization of opinions. We propose approaches tomeasuring the polarization of opinions and present simulation results. We show that the proposedpolarization index for a network allows one to distinguish and evaluate situations withmeaningfully different multidimensional distributions of opinions in a social network as well as tofind directions of the greatest polarization.

The spread of COVID-19 has forced governments to impose unprecedented restrictive measures on their people. Lockdowns have been introduced in many countries, and most governments have decided to require citizens to wear masks. The current study characterizes users’ attitudes towards the face mask requirements introduced by the Russian government as a response to the COVID-19 pandemic. We study how they relate to other users’ characteristics such as age, gender, and political attitudes. Subscriptions to information sources serve as a proxy for users’ political attitudes. Our results indicate that men and elder individuals—demographic groups that are most vulnerable to COVID-19—underestimate the benefits of wearing face masks, comparing to young people and women who demonstrate a higher rate of approval. We also discovered that users in opposition to the Russian government highly approve of this anti-COVID-19 measure, a result that is quite counterintuitive since the oppositionists should, a priori, negatively perceive all measures taken by the government. Further, we developed a probabilistic model that fixes the relationship between pure types (conservative/oppositionist;reject mask-wearing/approve of mask-wearing) and derived the method of estimating its parameters. We obtained that an oppositionist approves of the face mask requirements with the probability of 0.95. For those who support the Russian government, the odds of approval are merely 0.45. ©2021 ASSA.

This paper introduces a constructive definition of an informational community, which agrees with formal models of opinion dynamics for bounded rational agents in social networks. As is shown below, uncertainty can be taken into account when detecting informational communities. An example of a stable informational community is given. A control problem for informational communities is stated, and its solution is presented within the DeGroot model.

In this paper, we consider the problem of determining politico-ideological preferences of users of online social networks. We propose a (DLS) model that allows assessing politico-ideological preferences of VKontakte users using information from their accounts (digital footprints). The model meets the main ideological directions in modern-day Russia. Our approach is based on the supervised learning methodology whereby one solves a classification problem by calculating posterior probabilities of class membership. We compile an anonymized labeled dataset and develop appropriate software. Then we formulate and solve the problem of the politician’s choice of an ideological positioning strategy that will potentially receive the most support from the considered set of users. We exemplify our methodology by finding “ideal” political positioning for some popular online communities in VKontakte.

Many micro-level models of information processes in social networks consider either a change in the information-psychological state of agents (opinions, beliefs, attitudes) or a change in their observable behavior (actions). The goal in this paper is to develop and analyze a complex agent-based model of opinion dynamics that describes the dynamics of agents’ beliefs and the process of performing actions by agents. The issues of reaching consensus and polarizing opinions of agents are investigated.

In the context of the Coronavirus pandemic, it is crucial to analyze society's reaction to the information agenda related to the pandemic. This study examines news produced and broadcast by socially significant sources of information (in particular, accounts of news agencies, magazines, and newspapers) on the online social network VKontakte and Vkontakte users' consumption of these sources' news.