The rapid growth of digital communication platforms has generated vast volumes of user-generated textual data and digital footprints, creating growing demand for scalable artificial intelligence systems capable of supporting evidence-based decision-making. This study proposes and evaluates a human–AI collaborative analytical pipeline for multi-class sentiment and aggression analysis of large-scale social media data (N = 15,064 messages) related to an urban infrastructure project. The proposed framework integrates standard NLP preprocessing, machine learning-based classifiers, temporal aggregation, and controlled large language model (LLM)-assisted classification within a structured analytical workflow that incorporates expert validation and oversight. A stratified manual validation procedure (n = 301) demonstrated substantial inter-annotator agreement (κ = 0.70) and stable multi-class classification accuracy (80%). The results indicate that combining sentiment polarity and aggression detection as complementary linguistic indicators improves sensitivity to shifts in discourse dynamics and enables early identification of emerging social tension. The study demonstrates the potential of human–AI collaborative analytical frameworks for transparent, interpretable, and predictive large-scale social media analysis in decision-support contexts.

The paper deals with the analysis of the communicative behavior of various types of actors, speech perception and optimization of influence based on social media data and is an extended version of the report presented at CSoNet 2020 and published based on the deliverables of the conference. The paper proposes an improved methodology that is tested on the new material of conflicts regarding urban planning. The research was conducted on the material of social media concerning the construction of the South-East Chord in Moscow (Russia). The study involved a cross-disciplinary approach using neural network technologies, complex networks analysis. The dataset included social networks, microblogs, forums, blogs, videos, reviews. This paper presents the semantic model for the influence maximization analysis in social networks using neural network technologies, also proposed a variant of analyzing the situation with individual and collective actors, multiple opinion leaders, with a dynamic transformation of the hierarchy and ratings according to various parameters.

Industry 5.0 has transformed manufacturing with smart factories as a key component. This report reviews the latest research on smart factories and industrial data management. Focusing on automation, AI, machine learning, big data and virtualization Technologies such as IoT and crowd sensing enable continuous data collection for real-time decision making and human-machine collaboration. We present a smart factory architecture that combines edge computing for local processing and cloud analytics with various communication technologies. To increase safety, efficiency, and connectivity in modern factories

The analysis of large volumes of data collected from heterogeneous sources is increasingly important for the development of megacities, the advancement of smart city technologies, and ensuring a high quality of life for citizens. This study aimed to develop algorithms for analyzing and interpreting social media data to assess citizens’ opinions in real time and for verifying and examining data to analyze social tension and predict the development of situations during the implementation of urban projects. The developed algorithms were tested using an urban project in the field of transportation system development. The study’s material included data from social networks, messenger channels and chats, video hosting platforms, blogs, microblogs, forums, and review sites. An interdisciplinary approach was utilized to analyze the data, employing tools such as Brand Analytics, TextAnalyst 2.32, GPT-3.5, GPT-4, GPT-4o, and Tableau. The results of the data analysis showed identical outcomes, indicating a neutral perception among users and the absence of social tension surrounding the project’s implementation, allowing for the prediction of a calm development of the situation. Additionally, recommendations were developed to avert potential conflicts and eliminate sources of social tension for decision-making purposes.

The article considers the application of artificial intelligence technologies for successful implementation of projects launched in the urban transport system with regard to residents’ reactions. During the research, an algorithm was developed and subsequently tested on the material of urban transport conflicts. The study involved an interdisciplinary approach utilizing neural network technologies for real-time big data analysis. Data from social networks, microblogs, forums, blogs, video hosting sites, and reviews were used for the research. The developed algorithm helped identify conflict zones, determine topics and semantic accents that set off the strongest negative reactions from residents, and formulate recommendations to mitigate conflicts and create conditions for successful implementation of urban projects.

This study scrutinizes the enduring effects of racial and gender biases that contribute to the consistent underrepresentation of minority women in leadership roles within American private, public, and third sector organizations. We adopt a behavioural data science approach, merging psychological schema theory with sociological intersectionality theory, to evaluate the enduring implications of these biases on female leadership development using mixed methods including machine learning and econometric analysis. Our examination is concentrated on Black female leaders, employing an extensive analysis of leadership rhetoric data spanning 200 years across the aforementioned sectors. We shed light on the continued scarcity of minority female representation in leadership roles, highlighting the role of intersectionality dynamics. Despite Black female leaders frequently embracing higher risks to counter intersectional invisibility compared to their White counterparts, their aspirations are not realized and problems not solved generation after generation, forcing Black female leaders to concentrate on the same issues for dozens and, sometimes, hundreds of years. Our findings suggest that the compound influence of racial and gender biases hinders the advancement of minority female leadership by perpetuating stereotypical behavioral schemas, leading to persistent discriminatory outcomes. We argue for the necessity of organizations to initiate a cultural transformation that fosters positive experiences for future generations of female leaders, recommending a shift in focus from improving outcomes for specific groups to creating an inclusive leadership culture.

This research work focuses on the fake news classification in Bangla language using deep neural networks and machine learning classification algorithms processing the text data in the prominent way. Bangla language is the fifth most spoken native language in the world with approximately over 300 million native speakers and another 50 million as second language speakers. In this work, news collected from different online and print newspapers are classified in authentic and fake news class. Considering prominent natural language processing techniques, data preprocessing and performing different deep neural networks and machine learning classification algorithms, it has achieved a maximum of 81

Employee turnover (ET) is a major issue faced by firms in all business sectors. Artificial intelligence (AI) machine learning (ML) prediction models can help to classify the likelihood of employees voluntarily departing from employment using historical employee datasets. However, output responses generated by these AI-based ML models lack transparency and interpretability, making it difficult for HR managers to understand the rationale behind the AI predictions. If managers do not understand how and why responses are generated by AI models based on the input datasets, it is unlikely to augment data-driven decision-making and bring value to the organisations. The main purpose of this article is to demonstrate the capability of Local Interpretable Model-Agnostic Explanations (LIME) technique to intuitively explain the ET predictions generated by AI-based ML models for a given employee dataset to HR managers. From a theoretical perspective, we contribute to the International Human Resource Management literature by presenting a conceptual review of AI algorithmic transparency and then discussing its significance to sustain competitive advantage by using the principles of resource-based view theory. We also offer a transparent AI implementation framework using LIME which will provide a useful guide for HR managers to increase the explainability of the AI-based ML models, and therefore mitigate trust issues in data-driven decision-making.

The problem of recognizing and classifying emotions in speech is one of the most relevant and significant research topics, however, hardly any studies have been conducted to date for a large number of languages to achieve the required accuracy. Expressing and recognizing emotions based on the signal of the human speech is one of the complex issues that is distinct from languages. This paper proposes a systematical and robust approach to implement an emotion recognition system for low resource languages such as Persian. To the best of our knowledge, this is the first SER work on the Persian language using deep learning techniques. Sharif Emotional Speech Database ShEMO with five basic emotions including anger, fear, happiness, sadness and surprise, as well as neutral state is identified as suitable candidate to evaluate a 1D Convolutional Neural Network (1DCNN) architecture. The data are first processed using Mel-Frequency Cepstral Coefficients (MFCC) feature extraction method and then feed MFCC as input feature to our neural network. Experimental results demonstrate that our proposed method achieves about 74% classification accuracy on ShEMO dataset.

The article presents a methodology for analyzing the perception of a situation on the basis of digital data using the social stress and well-being indices. The material for the study is data from social networks, microblogs, forums, blogs, video hosting sites, reviews on the construction of road transport facilities in Moscow (South-East Chord, Southern Beltroad (Reconstruction of Verkhniye Polya Street between Maryinsky Park Street and the Moscow Ring Road), North-East Chord). The study involves a cross-disciplinary approach. To interpret the content, neural network text analysis, content analysis, sentiment analysis and analysis of lexical associations were used. Social media data made it possible to conduct real-time analytics, analyze users' perceptions and obtain a forecast on the further development of the situation around the analyzed urban development projects. The social stress and social well-being indices made it possible to characterize the communicative situation, taking into account the level of conflict potential and the dynamics of digital aggression, to identify the critical points in the conflict development and to determine the degree of the actors' satisfaction with the course of construction.

This chapter considers the link between productivity and wellbeing in the context of SMEs. The authors formulate a business–wellbeing–productivity framework, which connects business size and organisational structure with wellbeing parameters, which, in turn, are correlated with productivity. Using a country-level data set of Organisation for Economic Co-operation and Development (OECD) countries, we show that prevalence of SMEs in a country's business sector is associated with the decrease in productivity of this country through SMEs' negative impact on workforce wellbeing. Implications of this result for theory and practice are discussed.

The paper is concerned with information retrieval and analysis of digital conflictogenic zones based on social media data reflecting the users’ perception of road construction in Moscow. The material for the study was data from social networks, microblogs, blogs, instant messengers, videos, forums, and reviews dedicated to the construction of the South-East, North-East, and North-West Chords in Moscow. The study involved a transdisciplinary approach, neural network text analysis, content analysis, sentiment analysis, and analysis of word associations. The study made it possible to draw conclusions about the extremely tense situation around the construction of the South-East Chord. The level of aggression and social stress is quite high and approaching a critical point, and allows predicting further escalation of the conflict in the online and offline space. The construction of the North-East Chord causes some tension among the city residents and makes it possible to predict the development of a conflict in the virtual environment. The implementation of the North-West Chord project does not bear any special risks.

The article presents the results of the analysis of the adaptation of metropolis IT technologies to solve operational problems in extreme conditions during the COVID-19 pandemic. The material for the study was Russian-language data from social networks, microblogging, blogs, instant messengers, forums, reviews, video hosting services, thematic portals, online media, print media and TV related to the first wave of the COVID-19 pandemic in Russia. The data were collected between 1 March 2020 and 1 June 2020. The database size includes 85,493,717 characters. To analyze the content of social media, a multimodal approach was used involving neural network technologies, text analysis, sentiment-analysis and analysis of lexical associations. The transformation of old digital services and applications, as well as the emergence of new ones were analyzed in terms of the perception of digital communications by actors.

Much of business literature addresses the issues of consumer-centric design: how can businesses design customized services and products which accurately reflect consumer preferences? This paper uses data science natural language processing methodology to explore whether and to what extent emotions shape consumer preferences for media and entertainment content. Using a unique filtered dataset of 6,174 movie scripts, we generate a mapping of screen content to capture the emotional trajectory of each motion picture. We then combine the obtained mappings into clusters which represent groupings of consumer emotional journeys. These clusters are used to predict overall success parameters of the movies including box office revenues, viewer satisfaction levels (captured by IMDb ratings), awards, as well as the number of viewers' and critics' reviews. We find that like books all movie stories are dominated by 6 basic shapes. The highest box offices are associated with the Man in a Hole shape which is characterized by an emotional fall followed by an emotional rise. This shape results in financially successful movies irrespective of genre and production budget. Yet, Man in a Hole succeeds not because it produces most "liked" movies but because it generates most "talked about" movies. Interestingly, a carefully chosen combination of production budget and genre may produce a financially successful movie with any emotional shape. Implications of this analysis for generating on-demand content and for driving business model innovation in entertainment industries are discussed.

As it is known, the main difference between humans and animals is the presence of the second signaling system. The introduction of the second signaling system into the artificial intelligence system including the linguistic model of the world that works in conjunction with the extralinguistic model of the world, makes it possible to separate the presentation and processing of specific information into lower-level and upper-level representations. Thus, along with the detailed description, the model of the world also contains a generalized representation, which makes it easy for the user to interact with the system using natural language.

We propose a new systematic method to answer the research question: 'What is the financial and economic impact of servitizing the firm, digitising the firm, and combined servitization and digitization strategy?' Our method quantifies servitization, digitization and their synergy by analysing their relationship with firm financial and economic outcomes. The method is applied to the British publishing industry. Using text-mining and econometric analysis of secondary data, 258 UK book publishers (93% of the market share) are analysed over a period of 10 years (1,508 observations). Firms are categorised as servitized (S-firms), digitized (D-firms), digitized and servitized (DS-firms) and pure (P-firms) that are neither servitized nor digitized (control group). We detect no significant difference in terms of productivity and profitability between P-firms and D-firms. Although we find evidence of a servitization paradox, both S-firms and DS-firms show greater productivity than P-firms. Profitability of DS-firms is greater than that of P-firms, but profitability of S-firms is lower than that of P-firms. The research improves on the existing methodology employed to examine the impact of servitizing or digitising the firm and provides a means to measure how servitization and digitization impact on the productivity and profitability of a firm within a specific context.

The research is devoted to the analysis of hierarchies in inter-personal and intergroup network communication an example of a strategy of the election campaign in the Moscow City Duma in 2019. The study involved a cross-disciplinary approach using neural network technologies, complex networks analysis. For the correct interpretation of the content, content analysis, semantic analysis and analysis of word association were performed. The dataset included social networks, microblogs, forums, blogs, videos, reviews. The expansion and enrichment of the users' world view in the network environment in the analyzed communicative situation occurs through spreading of a more developed and well-founded model of the world, the carrier of which is a social media influencer with the necessary set of knowledge, techniques, a high level of some assets, who is able to communicate current requirements. Also the study made it possible to identify a level of social stress.

The purpose of this chapter is to provide an overview of how distributed ledger technologies ('blockchain') are currently being developed and deployed in the healthcare sector. Part I of the chapter provides an overview of data sources and data categories utilised in the study before analysing and highlighting the main findings of this study. Three findings are noteworthy: firstly, that the majority of blockchain for healthcare applications are still at the 'concept' or 'development' stage, and do not yet have an available product or service. Secondly, blockchain for healthcare applications that are promoted in the English language are predominately located in the United States. Thirdly, there are four observable categories of blockchain for healthcare applications with the most prevalent of these concerned with patient sovereignty over medical data. Part II of examines these four categories namely patient care, patient sovereignty over health data, healthcare administration and medical research in more detail. Part II of the chapter discusses the insights gleaned from the key findings of the study before some concluding remarks are made.

The structure of natural language could be considered a semantic network. This implies the allocation of the speech markers, which describe the subject and semantic areas. In this article, a wide range of texts about the digital economy was analyzed, making it possible to show the thematic structure of this subject area. Central and peripheral concepts were identified to characterize theoretical core concepts and related topics clarifying the application of the digital economy. Identification of the thematic areas was performed in two ways—through the construction of a thematic tree (neural network modeling in the Text Analyst) and the analysis of semantic networks. The results, approaches, and methods of this study could be used during the investigation of the other large thematic fields related to new ideological currents, being developed as an element of social design and management.