Search results

This study proposes a collaborative federated learning (CFL) framework to address personal data transmission and retention issues for artificial intelligence (AI)-enabled video surveillance in public spaces.

This study examines specific challenges for long-term people monitoring in public spaces and defines AI-enabled video surveillance requirements. Based on the requirements, this study proposes a CFL framework to gradually adapt AI models’ knowledge while reducing personal data transmission and retention. The framework uses three different federated learning strategies to rapidly learn from different new data sources while minimizing personal data transmission and retention to a central machine.

The findings confirm that the proposed CFL framework can help minimize the use of personal data without compromising the AI model's performance. The gradual learning strategies help develop AI-enabled video surveillance that continuously adapts for long-term deployment in public spaces.

This study makes two specific contributions to advance the development of AI-enabled video surveillance in public spaces. First, it examines specific challenges for long-term people monitoring in public spaces and defines AI-enabled video surveillance requirements. Second, it proposes a CFL framework to minimize data transmission and retention for AI-enabled video surveillance. The study provides comprehensive experimental results to evaluate the effectiveness of the proposed framework in the context of facial expression recognition (FER) which involves large-scale datasets.

This paper aims to explore the impact of green finance on the heterogeneity of enterprise green technology innovation and the underlying mechanism between them.

Using the data of China's A-share listed enterprises from 2008 to 2020 and the fixed effect model, the authors empirically explore the relationship and mechanism between green finance and green technology innovation by constructing the green finance index while considering both the quality and quantity of innovation.

The study suggests that green finance is positively related to the quality and quantity of enterprise green technology innovation, while green finance is more effective in stimulating the quality of green technology innovation than quantity. In addition, alleviating financial mismatch and improving the quality of environmental information disclosure are core mechanisms during the process of green finance facilitating green technology innovation. Furthermore, green finance exerts a more positive effect on the quality and quantity of green technology innovation with large-size enterprises, heavily polluting industries and enterprises in the eastern region.

This paper enriches the literature on green finance and green technology innovation and provides practical significance for green finance implementation.

The purpose of this study is to explore the impact of online reviews on answer adoption in virtual Q&A communities, with an eye toward extending knowledge exchange and community management.

Online reviews contain rich cognitive and emotional information about community members regarding the provided answers. As feedback information on answers, it is crucial to explore how online reviews affect answer adoption. Based on signaling theory, a research model reflecting the influence of online reviews on answer adoption is established and empirically examined by using secondary data with 69,597 Q&A data and user data collected from Zhihu. Meanwhile, the moderating effects of the informational and emotional consistency of reviews and answers are examined.

The negative binomial regression results show that both answer-related signals (informational support and emotional support) and answerers-related signals (answerers’ reputations and expertise) positively impact answer adoption. The informational consistency of reviews and answers negatively moderates the relationships among information support, emotional support and answer adoption but positively moderates the effect of answerers’ expertise on answer adoption. Furthermore, the emotional consistency of reviews and answers positively moderates the effect of information support and answerers’ reputations on answer adoption.

Although previous studies have investigated the impacts of answer content, answer source credibility and personal characteristics of knowledge seekers on answer adoption in virtual Q&A communities, few have examined the impact of online reviews on answer adoption. This study explores the impacts of informational and emotional feedback in online reviews on answer adoption from a signaling theory perspective. The results not only provide unique ideas for community managers to optimize community design and operation but also inspire community users to provide or utilize knowledge, thereby reducing knowledge search costs and improving knowledge exchange efficiency.

Due to the worldwide internet usage and online presence of businesses, digital marketing is an integrated part of the overall marketing strategy. One of the main investments in this field is the investment on more creative content that engages more potential customers. Online brand communities (OBCs) are used by businesses as a tool to have higher rates of engagement. Many researchers have developed customer engagement (CE) models to better understand it, but the transition economies like the countries of the Balkan are somehow left aside. Due to this identified gap in the literature, this paper aims to provide a model explaining the drivers and outcomes of CE, specifically on OBCs.

The research is based on primary data collected with an online questionnaire. Sample consisted of 764 respondents from four Balkan countries (Kosovo, North Macedonia, Albania and Montenegro). Data were analyzed using SPSS Statistics and IBM AMOS 26. To test the research model and hypotheses, confirmatory factor analysis and structural equation modeling were used.

First, the study found a positive impact of entertainment and customer satisfaction, and celebrities’ endorsement in CE. Furthermore, the results revealed a positive impact of CE on customer purchase intention and customer loyalty.

The research findings aim to help businesses of Balkan countries to better understand the importance and operation of CE in OBC.

This paper enriches the existing digital marketing literature of transition economies with a model of online CE.

The purpose of this paper is to provide a critical historical analysis of the business (mis)behaviors and influencing factors that discourage enduring cooperation between principals and agents, to introduce strategies that embrace the social values, economic motivation and institutional designs historically adopted to curtail dishonest acts in international business and to inform an improved principal–agent theory that reflects principal–agent reciprocity as shaped by social, political, cultural, economic, strategic and ideological forces

The critical historical research method is used to analyze Chinese compradors and the foreign companies they served in pre-1949 China.

Business practitioners can extend orthodox principal–agent theory by scrutinizing the complex interactions between local agents and foreign companies. Instead of agents pursuing their economic interests exclusively, as posited by principal–agent theory, they also may pursue principal-shared interests (as suggested by stewardship theory) because of social norms and cultural values that can affect business-related choices and the social bonds built between principals and agents.

The behaviors of compradors and foreign companies in pre-1949 China suggest international business practices for shaping social bonds between principals and agents and foreign principals’ creative efforts to enhance shared interests with local agents.

Understanding principal–agent theory’s limitations can help international management scholars and practitioners mitigate transaction partners’ dishonest acts.

A critical historical analysis of intermediary businesspeople’s (mis)behavior in pre-1949 (1840–1949) China can inform the generalizability of principal–agent theory and contemporary business strategies for minimizing agents’ dishonest acts.

The purpose of this study is to investigate the effects of celebrity endorsers on consumer advocacy, customization and entertainment intentions based on the notion of the stimulus–organism–response (SOR) framework. In addition, this study aims to examine the mediating and moderating role of customization and entertainment intentions on the proposed relationships. The authors also intend to highlight the comparative effects between male and female young social media consumers on those proposed associations.

In the pursuit of comprehensive and rigorous data collection, this study adopted a quantitative methodology using a meticulously crafted questionnaire. The questionnaire survey was conducted in major cities of Malaysia using the convenience and snowball sampling techniques. A total of 576 responses were collected, even though 549 retorts were used for data analysis. In this investigation, the authors strategically used covariance-based structural equation modeling through the use of AMOS v. 24 as the primary data analysis tool. Augmenting the analytical depth, the authors also conducted a supplementary bootstrap analysis. The additional layers of examination were crucial for appraising the mediating and moderating effects inherent within the model, in which the PROCESS MACRO v.4.20 was used.

The results of this study revealed the significant direct positive effects of celebrity endorsers on consumer customization, entertainment and advocacy intentions. Consumer customization and entertainment intentions also found significant direct affirmative effects on consumer advocacy intention, along with the significant direct positive effects of consumer entertainment intention on consumer customization intention. The results further revealed that consumer customization and entertainment intentions cannot mediate the relationship between celebrity endorsers and advocacy intention. The entertainment intention also declined the mediating effects between celebrity endorsers and consumer customization intention. On the contrary, consumer customization intention significantly and positively, and entertainment intention also significantly but negatively, moderate the association between celebrity endorsers and consumer advocacy intention. This study also illustrates that the effects of those examined relationships differ between male and female young social media consumers.

This study investigates the impact of celebrity endorsers on consumer behavior, focusing on their customization, entertainment and advocacy intentions. It extends current SOR framework, enhances source credibility theory, fills gaps in the literature on social media brand engagement and underscores the significance of customization and entertainment intentions. The findings provide insights for managers aiming to harness consumer brand advocacy through celebrity endorsers effectively.

This paper seeks to address the following two research questions. RQ1: What are the influential user archetypes in the social question-answering (SQA) community? RQ2: To what extent does user feedback affect influential users in changing from one archetype to another?

Based on a sample of 13,840 influential users drawn from the Covid-19 community on Zhihu, the archetypes of influential users were derived from their ongoing participation behavior in the community using the Gaussian mixture model. Additionally, user feedback characteristics such as relevance and volume from 222,965 commenters who contributed 546,344 comments were analyzed using the multinomial logistic regression model to investigate the archetype change of influential users.

Findings suggest that influential users could be clustered into three distinctive archetypes: touch-and-go influential users, proactive influential users and super influential users. Moreover, feedback variables have various impacts on the influential user archetype change, including a shift toward creating higher-quality content and fostering increased interaction, a shift toward generating lower-quality content and decreased interaction but improved speed and having mixed effects due to differences in information processing among these archetypes.

This study expands the existing knowledge of influential users and proposes practical approaches to cultivate them further.

The purpose of this study was to explicate how leaders’ knowledge hiding results in employees’ knowledge hiding. In addition, the study was intended to explore under what conditions leaders’ knowledge hiding affects employees’ moral disengagement more deleteriously.

Data were collected from 321 employees at three different times which were two months apart from each other. Structural equation modeling was used for data analysis.

The study found leaders’ knowledge hiding to be related to employee moral disengagement. In addition, the study found moral disengagement to affect employees’ knowledge-hiding behavior. Moral disengagement was found to mediate the relationship between leaders’ knowledge hiding and employees’ knowledge hiding. Finally, the study found that employees with high moral identity show more perseverance to preserve their moral engagement when led by knowledge-hiding leaders.

To the best of the authors’ knowledge, the study was first to establish a relationship between a leader’s knowledge hiding and employees’ moral disengagement. The study also established the mediating role of moral disengagement to work as a mediating mechanism linking leaders’ knowledge hiding to employees’ knowledge hiding. Finally, the study found that moral identity moderates the relationship between leaders’ knowledge hiding and employees’ moral disengagement.

Public service announcements (PSAs) have been shown to be effective instruments that raise awareness, educate society, and change behaviors and attitudes. Many governments and organizations have utilized PSAs on social media to promote online safety among children and youth. However, we have limited understanding of the range of topics that these PSAs address and how they present their content to audiences. This study provides an inventory of the types of online safety topics that current PSAs address and a catalogue of the types of persuasive features employed by PSAs.

A content analysis of 220 YouTube PSA videos on online safety was conducted. Various topics under the umbrella of online safety were identified. Guided by the prospect theory and exemplification theory, different persuasive features employed in the PSAs were sought.

The findings highlight that the primary focus of these PSAs is on online safety behaviors and general instructions on online hygiene. Interestingly, nearly half of the videos employ a neutral frame, while a significant portion provides no evidential support. Additionally, video length was associated with the number of views and likes it gathered but not with the number of comments.

The inventory of PSAs can help researchers, practitioners, and policymakers better understand the type of content being produced and disseminated online as well as identify topics that are either over or under-represented. Further, the catalogue of the types of persuasive features employed by PSAs would be helpful in guiding research, practice, and policymaking in the context of creating effective online safety videos.

Knowledge adoption is the key to effective knowledge exchange in virtual question-and-answer (Q&A) communities. Although previous studies have examined the effects of knowledge content, knowledge source credibility and the personal characteristics of knowledge seekers on knowledge adoption in virtual Q&A communities from a static perspective, the impact of answer deviation on knowledge adoption has rarely been explored from a context-based perspective. The purpose of this study is to explore the impact of two-way deviation on knowledge adoption in virtual Q&A communities, with the aim of expanding the understanding of knowledge exchange and community management.

The same question and the same answerer often yield multiple answers. Knowledge seekers usually read multiple answers to make adoption decisions. The impact of deviations among answers on knowledge seekers' knowledge adoption is critical. From a context-based perspective, a research model of the impact of the deviation of horizontal and vertical answers on knowledge adoption is established based on the heuristic-systematic model (HSM) and empirically examined with 88,287 Q&A data points and answerer data collected from Zhihu. Additionally, the moderation effects of static factors such as answerer reputation and answer length are examined.

The negative binomial regression results show that the content and emotion deviation of horizontal answers negatively affect knowledge seekers' knowledge adoption. The content deviation of vertical answers is negatively associated with knowledge adoption, while the emotion deviation of vertical answers is positively related to knowledge adoption. Moreover, answerer reputation positively moderates the negative effect of the emotion deviation of horizontal answers on knowledge adoption. Answer length weakens the negative correlation between the content deviation of horizontal and vertical answers and knowledge adoption.

This study extends previous research on knowledge adoption from a static perspective to a context-based perspective. Moreover, information deviation is expanded from a one-way variable to a two-way variable. The combined effects of static and contextual factors on knowledge adoption are further uncovered. This study can not only help knowledge seekers identify the best answers but also help virtual Q&A community managers optimize community design and operation to reduce the cost of knowledge search and improve the efficiency of knowledge exchange.