Abstract Chatbots are increasingly part of digital customer service, yet their evaluation is often reduced to response speed, availability, or automation rate. This study assumes that customers do not experience the algorithm directly, but the language through which the system communicates. The empirical analysis is based on survey data from 411 respondents. Composite indices were constructed for language quality, trust, satisfaction, service communication performance, and customer re-engagement intention, all showing acceptable internal consistency. Regression models indicate that language quality is positively associated with trust ( $$\:\beta\:=0.283;p<0.001$$ ). Language quality and trust jointly explain 50.7% of the variance in satisfaction. Customer re-engagement intention is most closely linked to satisfaction ( $$\:\beta\:=0.287;p<0.001$$ ), while the direct relationship between trust and re-engagement becomes non-significant after satisfaction is included. Mediation results support a sequential mechanism in which language quality is associated with re-engagement via trust and satisfaction. The study shifts chatbot evaluation from technical efficiency toward language quality as part of managing digital service operations.
This study examines the determinants of repurchase intention in the beauty clinic industry by developing an integrated framework linking perceived service quality, perceived price fairness, and influencer credibility with brand trust, and by testing brand trust as a mediating mechanism. Survey data were collected from 360 beauty clinic consumers in the Jabodetabek metropolitan area, Indonesia, using purposive sampling. The proposed model was tested using covariance-based structural equation modelling (SEM) with AMOS, with bootstrapping used to assess the mediating role of brand trust. Perceived service quality, perceived price fairness, and influencer credibility each had significant positive effects on brand trust and repurchase intention (all p < 0.05). Perceived service quality was the strongest predictor of brand trust (β = 0.399), while perceived price fairness was the strongest direct predictor of repurchase intention (β = 0.332). Brand trust significantly influenced repurchase intention (β = 0.296) and partially mediated all three antecedents, most strongly for perceived service quality (indirect β = 0.117). The model explained 30.6
Abstract This paper investigates an efficient approach for distilling Large Language Models (LLMs) into smaller, application-specific models using zero-shot Chain of Thought (CoT) rationale generation and Optimization by Prompting (OPRO). To address the challenges of deploying computationally intensive generative AI for narrow tasks or resource-constrained environments, the approach leverages LLM reasoning capabilities to generate both labels and natural language explanations for unlabeled data. By reducing reliance on human-generated annotations, the approach substantially lowers annotation requirements and prompting costs while maintaining comparable performance in the evaluated settings. We formulate distillation as a multi-task learning problem in which student models are trained to jointly predict labels and learn from teacher-generated rationales, with the goal of improving data efficiency and generalization. Building on established zero-shot Chain of Thought (CoT) prompting and the OPRO prompt optimization technique, we use teacher-generated rationales to reduce annotation token requirements and examine the associated performance and efficiency gains. Additionally, we systematically investigate how explanation properties affect distillation efficiency. Across natural language inference and question answering benchmarks, results indicate that near-optimal performance can be achieved even when rationales are provided for only a subset of the training data, and that shorter explanations are often sufficient. These findings provide practical insights into the trade-offs between rationale generation cost and student model performance. Overall, this work contributes empirical evidence on the effectiveness and cost characteristics of rationale-based distillation for training compact, task-specific language models with minimal human intervention.
While existing literature establishes that patent quality drives firm performance better than mere patent counts, the boundary conditions of this consensus remain underexplored under severe macroeconomic stress. This study investigates the relationship between patent quantity and quality and firm profitability – in terms of return on assets (ROA) and return on equity (ROE) – within the strategically vulnerable, capital-intensive Western European electronics manufacturing sector over the 2016–2024 period. Using the European Patent Office’s PATSTAT database and EMIS, we analyse a panel of 112 firms via two-way fixed-effects regressions to test how extreme economic disruptions impact innovation premiums. Our baseline findings confirm that under stable macroeconomic conditions, patent quality – measured by average forward citations – acts as a robust exogenous driver of operating and equity profitability, whereas sheer patent quantity does not. Crucially, however, we demonstrate that the systemic post-2020 economic shock and subsequent disruptions completely neutralised this established quality premium. This study provides a novel theoretical contribution by showing that the superiority of high-quality intellectual property is not absolute; rather, it is strictly contingent upon stable macroeconomic environments. By highlighting these vulnerabilities, our findings offer critical insights into the limitations of patent-driven profitability during periods of systemic economic uncertainty.
The article examines the relationship between population body dimensions and the practical design of wooden furniture, with a particular focus on wooden chairs intended for bariatric individuals. Given the growing prevalence of obesity, there is an increasing need to reassess the ergonomic, dimensional, and strength requirements of seating furniture for this population group. The aim of the study is to analyse selected body dimensions of bariatric respondents in Slovakia and to identify their opinions and preferences regarding seating furniture designed for overweight users. The research was based on a comprehensive analysis of selected body dimensions of 319 bariatric respondents in Slovakia. In addition, a structured questionnaire survey was conducted to assess 350 respondents’ preferences and their level of satisfaction with the existing portfolio of wooden seating furniture. The findings indicate that the current portfolio of wooden furniture does not sufficiently meet the needs of bariatric users. Respondents emphasized the need for improved chair dimensions and greater structural strength. The results also show consumer interest in furniture adapted to above-average body weight and dimensions. The study confirms the necessity of redesigning wooden seating furniture to better accommodate bariatric individuals. Special attention should be paid to ergonomic parameters, load-bearing capacity, and appropriate dimensions in order to increase user comfort, safety, and satisfaction.
The global olive oil sector is currently transitioning from mass production toward high-value differentiation. This research explores the strategic shift in extra virgin olive oil (EVOO) positioning through the creation of a distinct market niche within the luxury sector, contributing a framework for Business Model Innovation that enhances consumer value. A qualitative case study focuses on a Greek Small and Medium-sized Enterprise (SME) implementing a luxury positioning strategy. Primary data were collected through in-depth semi-structured interviews with the firm’s leadership and direct observation. Data triangulation included secondary data and external validation through feedback from key stakeholders. Operational tools such as the Threats-Opportunities-Weaknesses-Strengths (TOWS) matrix and the strategy map architecture were used strictly as management support instruments to visualize and execute the transformation of the business model under consideration. The results show that shifting toward a luxury food model requires combining superior product quality anchored in certified organoleptic characteristics with strategic branding and artification. This study highlights that an SME adopting an asset-light model and focusing on orchestrating the stakeholders involved in the value chain can achieve prestige pricing. The emphasis on a distinct sensory profile creates a complete experience and emotionally engages consumers. The study offers a replicable strategic model for food and beverage entrepreneurs who wish to move from the classic commodity model toward an elevated market positioning.
Digital transformation has become a strategic imperative across Europe, yet significant disparities persist between member states and within national economies. This study examines the determinants of digital maturity in Romanian enterprises using data from 199 firms assessed through the national Digital Maturity Assessment (DMA). By integrating regional, sectoral, and organizational characteristics with capability-based indicators, the analysis shows that Romania remains a developing digital ecosystem marked by low levels of automation and artificial intelligence (AI) adoption, persistent regional disparities, and sectoral fragmentation. Multiple regression results indicate that Human-Centric Digitalization, Automation AI readiness, and Green Digitalization are the capability domains most strongly associated with overall digital maturity, collectively explaining more than 91
Virtual reality (VR) has emerged as a strategic tool for destination marketing; however, the mechanisms through which VR-based authentic experiences translate into revisit intention remain insufficiently theorized. Drawing upon multidimensional authenticity theory and relationship marketing perspectives, this study develops and empirically tests a structural model explaining how VR authentic experience influences revisit intention through differentiated authenticity perceptions, relational orientation, and advocacy mechanisms. Using survey data from 857 respondents and PLS-SEM analysis, the findings demonstrate that VR authentic experience significantly activates all four authenticity dimensions. Existential authenticity (β = 0.481) and constructive authenticity (β = 0.479) exhibit nearly identical and strongest effects, followed by objective authenticity (β = 0.423) and postmodern authenticity (β = 0.317). Results reveal that authenticity perceptions primarily influence revisit intention indirectly through perceived destination brand relationship orientation (PDBRO), which emerged as the most powerful direct predictor of revisit intention (β = 0.574). Mediation analysis confirms that relational orientation constitutes the dominant transmission mechanism, particularly for constructive and objective authenticity, whereas advocacy plays a comparatively weaker role. Furthermore, destination preference selectively moderates the advocacy-revisit relationship but does not strengthen the relational pathway, indicating that relational embeddedness operates independently of comparative prioritization. Overall, the findings reposition relational consolidation as the central mechanism through which VR-based authentic experiences generate future visitation intentions. By integrating authenticity theory with relationship marketing in immersive digital contexts, this study advances theoretical understanding of VR tourism and provides strategic insights for destination managers seeking to convert virtual engagement into sustained behavioral commitment.
Owing to the rapid development of programmatic advertising, the number of fraudulent activities has increased, thereby reducing the performance of digital marketing campaigns. Machine learning (ML)-based fraud detection systems have been identified as a solution to such fraud and to enhance advertising performance. This study examines the effects of ML-based fraud detection on the most important ad performance parameters, real-time bidding (RTB) efficiency, and consumer trust. This study utilizes information obtained from 780 marketing professionals to investigate the direct and indirect impacts of fraud detection systems on ad performance. The findings indicate that ML-based fraud detection positively impacts ad performance, such as click-through rates (CTR) and return on ad spend (ROAS), by improving RTB efficiency and decreasing fraud-induced distortions. Nonetheless, consumer trust, although positively influenced by fraud detection, has a less positive correlation with advertisement performance than anticipated. This study also analyzes how data privacy regulations moderate the relationship and increase the efficiency of RTB and consumer trust. The results can be useful to practitioners in the digital advertising field, as they show that it is crucial to combine fraud detection systems and follow the rules of data privacy to make ads more effective. This research adds to the theoretical background of the role of fraud detection in programmatic advertising and provides viable suggestions on how digital advertising results can be enhanced.
This article examines the economic resilience of Romania’s postal and courier sector in a multi-hazard context marked by the COVID-19 pandemic and the escalation of the Russia-Ukraine war. The sector is defined as postal activities under universal service obligation (NACE 5310) and other postal and courier activities (NACE 5320). Using annual county-level data for all 42 Romanian counties for 2008–2023, sourced from ListăFirme.ro and based on figures reported to the Ministry of Finance, the study combines bibliometric mapping with static panel regression using county fixed effects. The results show a positive association at the 5
Anomaly detection is widely used for monitoring complex systems, yet most surveys and benchmarks emphasize cybersecurity and financial fraud as common use cases, and leave other application areas under-mapped. This paper uses a bibliometric and topic modelling approach to chart recent anomaly detection research beyond these two dominant domains. We analyze 4,362 Web of Science articles published between 2022-2025 on anomaly detection, filtered to exclude finance- and security- related work. A Latent Dirichlet Allocation uncovers 12 topics, interpreted as nine application-oriented domains and three cross-domain, general topics. The application topics span behavioral and environmental pattern anomalies, energy systems faults and maintenance, process control and industrial operations, IoT edge–cloud and graph monitoring, industrial visual inspection, healthcare and biomedical diagnosis, hyperspectral remote sensing, and video surveillance. Topic prevalence and temporal trends show that behavioural and environmental patterns constitute the broadest cross-domain theme, while industrial visual inspection is the fastest-growing frontier over the analyzed period. Finally, by relating topics to Web of Science categories and to clusters of strongly coupled journals, the study identifies the main disciplinary anchors and publication venues that structure non-financial, non-cybersecurity anomaly detection and highlights emerging opportunities for cross-domain benchmarking and collaboration.
Exploration of factors driving consumer purchase intention growth in marketing practices is attracting increasing attention and continues to develop in recent studies. However, there is still a gap in the literature, especially studies on market orientation as a guideline that provides clarity of marketing direction that bridges the relationship between marketers' creative innovation and consumer purchase intention for SME products in tourism areas. Therefore, this study aims to examine the mediating effect of market orientation (MO) on the predictive relationship between marketers' creative innovation (MCI) and consumer purchase intention (CPI) in SMEs in geopark tourism areas in the Indonesian context. Quantitatively, an explanatory survey method was developed through a research instrument that successfully collected 118 respondents' responses, which were analyzed using Partial Least Squares-Structural Equation Modeling (PLS-SEM). The research findings prove that MCI has a direct positive effect on CPI. MCI also has a positive effect on MO, and MO is confirmed to have a positive effect on CPI. These findings also provide evidence that MO has a positive partial mediation effect on the relationship between MCI and CPI in SME products in geopark tourism areas. These findings practically highlight the importance of clarity and policy regarding destination direction, target customers, and competitor behavior information, which are crucial for facilitating marketers' creative innovation mechanisms as strategic competencies that can be developed within marketing practices to increase tourists' purchase intentions. This can be even more effective when integrated with artificial intelligence (AI)-based digital platforms.
This study explores the impact of artificial intelligence (AI) on the labour market, focusing on changes in job roles, skill requirements, and human resource (HR) practices. Unlike previous surveys that primarily addressed technological aspects, this research systematically integrates technological, organisational, and institutional perspectives. The literature review (2020–2025) shows that AI adoption is closely associated with rising demand for technical and interdisciplinary skills, restructuring of work roles, and widening wage gaps between AI-skilled and non-AI-skilled workers. At the same time, many jobs are not eliminated but transformed, as AI frequently functions in augmentation rather than pure substitution mode. The results suggest that the future of work in the AI era will be determined less by technology itself than by how organisations, institutions, and policies govern its implementation.
Employee resilience refers to a developable capacity to bounce back from adversity, to adapt and even thrive in response to challenging events. As a result of economic, social and environmental challenges organizations are facing and given the turbulent work environment, interest in employee resilience has been increasing, leading to a growth in the number of publications. The aim of this paper is to reveal how this literature on employee resilience is currently structured. For this, a bibliometric analysis of documents collected from Scopus database, including performance analysis, co-citation analysis, and co-word analysis, was conducted. Such analysis provided an overview of the construct and allowed to identify thematic clusters. After that, a qualitative analysis of TOP20 most cited articles was carried out to define future research avenues. The results revealed the existence of five thematic clusters: (1) employee resilience as a capacity of positive organizational behaviour and dimension of psychological capital (green cluster); (2) Job Demands-Resource theory as a lens for understanding employee resilience (purple cluster); (3) Conservation of Resources theory as a lens for understanding employee resilience (red cluster); (4) leadership as a driver of employee resilience (blue cluster); and (5) deeper disclosure of employee resilience: the fundamental question requires further elaboration (yellow cluster). Turning to future research avenues, seven were disclosed: digital transformation and employee resilience; demographics shaping employee resilience; improvements in research design and measurement; bringing employee resilience into the domain of HRM; linking employee resilience with organizational resilience; antecedents of employee resilience; and outcomes of employee resilience. The current paper is expected to guide a new wave of research by providing an overview of the current status of the knowledge base.
This study advances understanding of how Dynamic Capabilities (DC) influence Firm Performance (FP) among Small and Medium-Sized Enterprises (SMEs) in emerging economies, focusing on Ghana as a developing market context. While prior research has examined DC-FP relationships in advanced economies, this study uniquely explores the moderating role of government support in shaping this linkage within a resource-constrained environment. Using purposive sampling, data were collected through a three-phase approach involving online surveys, field interviews, and follow-up assessments, generating 298 valid responses from SME owners and managers. Structural Equation Modeling results reveal that sensing capabilities significantly enhance firm performance, while government support strengthens the positive relationship between integrating capabilities and firm performance. These findings underscore the critical role of government interventions in amplifying firms' capacity to integrate knowledge and resources for improved outcomes. The study contributes to the dynamic capabilities literature by contextualizing the DC-FP nexus in an African SME setting and offers managerial insights into leveraging and integrating capabilities by sharing unique ideas across business units to enhance firm performance in turbulent environments.
Besides ethical and privacy issues, Artificial intelligence (AI) also raises concerns from an environmental perspective. Training AI models requires very large datasets and entails substantial energy and water consumption (WC). Some projections suggest that, by 2027, the annual global energy demand attributable to AI could reach 85-134 TWh, while WC may amount to 4.2-6.6 billion cubic meters, equivalent to four to six times Denmark's annual consumption and nearly half of the United Kingdom's. These figures underscore the importance of assessing the sustainability of AI by considering its impacts across the entire life cycle. In this study, the environmental impacts of Generative AI (GenAI) were quantitatively assessed using the Life cycle assessment methodology, across 18 impact categories, based on secondary data. The results indicate that training a single GenAI model over q year could generate 767,814 kg CO2 eq (comparable to the annual emissions of 167 cars), 190,145 kBq Co-60 eq, 13.283 kg 1.4-DCB eq (equivalent to the pesticide treatment of approximately 102 ha of agricultural land), 28,485 m2a crop eq (about 4 football fields), and 184,690 kg oil eq (equivalent to burning roughly 615 barrels of oil). Overall, the findings highlight the considerable environmental burden of AI, with potential negative consequences comparable to those of entire polluting industrial sectors. It should be noted, however, that these estimates are conservative and intended for illustrative purposes. Actual resource consumption may be higher, depending on factors such as data center efficiency, the energy mix employed, and specific operating conditions.
The aim of the presented research was to quantitatively assess the extent to which behavioral factors (travel experience and autonomy in travel) and psychological factors (perceived travel stress and social discomfort) influence the desire to travel and explore new places among students belonging to Generation Z (Gen Z). The research was conducted in October 2024 on a sample of 404 respondents from the Alexander Dubcek University of Trencin, with the target group being students at the bachelor's and master's levels of study. Data collection was carried out through an online questionnaire, the content of which focused on four key areas: travel experience, autonomy in travel, perception of travel-related stress, and social discomfort during interaction in a foreign environment. The data analysis was conducted using correlation and multiple linear regression analysis. The results showed that the most significant positive predictor of the desire to travel was the number of countries visited (B = 0.1927; p < 0.01), while the perception of travel as a stressful factor had a statistically significant negative impact (B = -0.1563; p < 0.001). Other variables - willingness to travel alone and discomfort when ordering food - did not show statistical significance. Nonetheless, the model as a whole was statistically significant (F = 8.58; p < 0.001). These findings confirm the importance of behavioral and psychological factors in shaping travel motivation among young people. The research provides practical recommendations - such as strengthening students' travel competencies through international mobility and simultaneously reducing psychological barriers through the development of adaptation strategies. The results also provide a foundation for further research that could analyze a broader range of personality, value-based, and socio-environmental determinants of Gen Z's travel behavior.
This article examines the knowledge and abilities required by local leaders and professionals to effectively handle the Just Energy Transition (JT) in a disadvantaged European region. The research is based on qualitative analysis performed in the Hunedoara region, Romania, one of the regions anticipated to be most impacted by this change. Although the analysis focuses on Hunedoara, the challenges observed reflect patterns in coal-dependent regions in Europe. The focus is directed towards the important skills and knowledge required for change management, with approaches to communication critical for engaging diverse social stakeholders. Based on insights from three focus groups with 20 local participants (September 2023), the article underlines the importance of collaborative decision-making and the adjustment of communication methods to align with local contexts. The findings indicate a need to enhance leadership and administrative competencies, as well as ongoing deficiencies in comprehending the practical implications of the JT at the community level. The research illustrates the essential importance of public support and good communication in promoting knowledge, understanding, and acceptance of the Just Energy Transition, while providing insights into how local stakeholders might more effectively convey and execute the idea.
Our study aims to map the interdisciplinary research landscape encompassing artificial intelligence, environmental science, and economics through bibliometric analysis. By using the Web of Science academic database to retrieve publications based on the most representative words from the taxonomy of each concept, we applied citation analysis, co-authorship network mapping, and keyword co-occurrence analysis using the R package bibliometrix and VOS Viewer software to assess the growth, key contributors, collaboration patterns, and thematic focuses within this multidisciplinary field. Topics including “artificial intelligence,” “machine learning,” and “deep learning” for artificial intelligence (AI), “environmental footprint,” “sustainability,” and “climate change” for life sciences, and “economic impact,” “resource allocation,” and “business impact” for economics have been used as part of PRISMA methodology to narrow down the most relevant studies for the analysis. Additionally, database coverage comparison and word mining over definitions and taxonomy have been performed as a prerequisite for a robust dataset. The results revealed a significant increase in publications over the past decade, highlighting the expanding role of AI in tackling environmental and economic challenges. Keyword analysis exposed dominant themes such as sustainability, resource management, and economic impact assessment, alongside emerging trends in AI applications for environmental conservation. Moreover, the network mapping revealed the AI techniques most applied in environmental economics research, such as predictive modeling, optimization algorithms, and data analytics. Therefore, our findings underscore the critical integration of AI with environmental science and economics, revealing a dynamic and rapidly evolving research landscape. The study highlights the necessity of interdisciplinary collaboration in leveraging AI technologies for sustainable development and economic optimization. It offers a roadmap for future research, suggesting areas where AI can significantly contribute to addressing complex global challenges, with topic partitioning and evolution over time. This bibliometric study serves as an important resource for researchers and policymakers, guiding the development of integrated strategies that involve AI’s potential in the presented domains.
