Search results

This study aims to investigate the differences in consumer reviews across multiple e-commerce platforms to better assist consumers in making informed decisions. By examining the specific content of these differentiated reviews, the study seeks to provide insights that can enhance e-commerce services and improve consumer satisfaction.

The research utilizes the latent Dirichlet allocation (LDA) method for text analysis to identify the varying concerns of consumers across different e-commerce platforms for the same product. Additionally, the study expands the sentiment dictionary to address polysemy issues, allowing for a more precise capture of sentiment differences among consumers. A non-parametric test is employed to compare reviews across multiple platforms, providing a comprehensive analysis of review disparities.

The findings reveal that consumer concerns and sentiments vary significantly across different e-commerce platforms, even for the same product. The combination of text analysis and non-parametric testing highlights the objectivity of the research, offering valuable evidence and recommendations for improving e-commerce services and enhancing the shopping experience.

This study is original in its approach to combining text analysis with non-parametric testing to examine multi-platform review differences. The research not only contributes to the understanding of consumer behavior in the context of e-commerce but also provides practical suggestions for platforms and consumers, aiming to optimize service quality and consumer satisfaction.

In social media context, service failures become public domain, making them visible to vast audiences of customers who are virtually present. Thus, this study aims to discuss the roles of service failure type and management response on observers’ perceived helpfulness.

The authors conducted econometric analyses on a dataset incorporating 107,984 reviews and 34,641 management responses to negative reviews.

The results reveal that, for process failures, specifying a form of action (initiatives for solving the problem) is associated with more perceived helpfulness of reviews than accounts (explanation) or acknowledgments (recognition and acceptance), while responding to an outcome failure by providing an account is associated with increased perceived helpfulness of reviews.

For process failures, managers should make every effort to convince observers through specific actions that similar failures are less likely to occur in the future. For outcome failures, managers should strive to provide clear and fast explanations of the failure causes to restore observers’ evaluation of the firm’s capabilities.

The authors’ work extends sparse insights on observers and sheds new light on the effect of service failure type and response strategy on observers’ actual behaviors. The interplay between service failure type and response strategy provides guidance on how to use management responses to influence potential customers.

The continuing development of digital technologies creates expanding opportunities for information transparency. Consumers use social media to provide online reviews that are focused on changing levels of consumer trust. This study examines the effect of perceived risk that prompts consumers to search for online reviews in the context of food safety.

Commitment-trust theory forms the theoretical lens to model changes in consumer trust resulting from online reviews. Consumer-based questionnaire surveys collected data to test the structural model, using structural equation modelling (SEM).

The findings show when consumers perceive high levels of risk, they use social media to obtain additional product-related information. The objective, unanimous, evidential and noticeable online reviews are perceived as informative to consumers. Perceived informativeness of positive online reviews is found to increase consumers trust and, in turn, increase their purchase intentions.

The findings contribute to the knowledge of online review-based trust literature and provide far-reaching implications for information system (IS)-practitioners in business.

Given the popularity of online health communities (OHCs) and medical question-and-answer (Q&A) services, it is increasingly important to understand what constitutes useful answers and user-adopted standards in healthcare domain. However, few studies provide insights into how health information characteristics, provider characteristics and recipient characteristics jointly influence user information adoption decisions. To fill this research gap, this study examines the combined effects of physicians' certainty tone as information characteristics, seniority as provider characteristics and disease severity as recipient characteristics on patients' health information adoption.

Drawing on dual-process theory and information adoption model, an extended information adoption model is established in this study to examine the effect of attitude certainty on patients' health information adoption, and the moderating effects of online seniority and offline seniority, as well as patient motivation level—disease severity. Utilizing logit regression models, the authors empirically tested the hypotheses based on 4,224 Q&A records from a popular Chinese OHC.

The results show that (1) attitude certainty has a significant positive impact on patients' health information adoption, (2) the relationship between attitude certainty and information adoption is negatively moderated by physicians' online seniority, but is positively moderated by offline seniority; (3) there is a negative three-way interaction effect of attitude certainty, online seniority and disease severity on patients' health information adoption.

This study extends the information adoption model to examine the two-way interaction between argument quality and source reliability, as well as the three-way interaction with user motivation level, especially for health information adoption in the healthcare field. These findings also provide direct practical applications for knowledge contributors and OHCs.

Information filtering systems serve as robust tools in the ongoing difficulties associated with overwhelming volumes of data. With constant generation and accumulation of reviews in online communities, the ability to distill and provide valuable insights to assist customers in their search for relevant information is of considerable significance. This study devised an effective review filtering system for a popular online physical experience review site.

This study entailed an investigation of a hybrid approach for a review filtering system augmented with various text mining-based operational variables to extract the linguistic signals of online reviews. Moreover, we devised three ensemble models based on multiple machine learning and deep learning algorithms to build a high-performance review filtering system.

The main findings confirm the effectiveness of using the derived operational variables when reviewing filtering systems. We found that the reviewer’s tendency and history macros, as well as the readability and sentiment of the reviews, contribute significantly to the filtering performance. Furthermore, the proposed three ensemble frameworks demonstrated good efficiency with an average accuracy of 89.39%.

This study provides a methodological blueprint for operationalizing variables in online reviews, covering both structured and unstructured datasets. Incorporating different variables enhances the efficiency of the algorithm and provides a more comprehensive understanding of user-generated content. Furthermore, the study affords a strategic perspective and integrated guidelines for developers seeking to create advanced review filtering systems.

Language styles of online reviews are becoming increasingly important in consumers’ purchase decisions. However, there are inconsistencies in research on the effects of literal and figurative language styles in online reviews on service consumption. Drawing upon construal level theory, this research explores the effects of literal and figurative online reviews on consumers’ word-of-mouth recommendations and their internal mechanisms in the context of service consumption. In addition, this research identifies service types (experience vs credence services) as boundary conditions under which online review language styles play a role.

Three studies are designed to verify the effect of language style in online reviews on consumer word-of-mouth recommendations. Study 1 (N = 195) tests the interaction between construal level and (literal vs figurative) language style on consumers’ word-of-mouth recommendations. Study 2 (N = 191) identifies the depth of information processing as an underlying mechanism. Study 3 (N = 466) examines the boundary condition due to service type. The main methods used are independent sample t-test, ANOVA and bootstrapping.

The results illustrate that (1) consumers at different construal levels prefer online reviews with different language styles, and this can influence their word-of-mouth recommendations: consumers with a low construal level prefer online reviews with a literal language style, while those with a high construal level prefer online reviews with a figurative language style; (2) the depth of information processing plays a mediating role in the above interaction effect and (3) service type serves as a boundary condition such that the preference for literal (vs figurative) language style among low- (vs high-) construal-level consumers holds only for experience services; for credence services, online reviews with a literal language style enhance word-of-mouth recommendations, regardless of consumers’ construal level. The findings shed light on the drivers of word-of-mouth recommendations and provide insights to promote more effective word-of-mouth recommendations.

Drawing upon the construal level theory, this research explores the factors that influence online review language styles on consumer word-of-mouth recommendations and their underlying mechanisms and discusses the moderating effects of different service types (i.e. experience services and trust services). It not only sheds light on the contradictions in the previous literature but also provides new insights for academics and business managers to deepen their understanding of facilitating word-of-mouth recommendations.

This study proposes and tests a model to analyse whether achievement, social and immersion motivational affordances embedded in gamified review platforms motivate consumers to altruistically create content in the post-consumption stage.

We used data from a sample of 343 reviewers and employed SmartPLS to test the research model.

Findings revealed that, while achievement affordances (i.e. points, levels and badges) have no significant effect, immersion affordances (i.e. avatars) and more especially, social affordances (i.e. receiving helpful votes from readers and having followers) are key for review platforms, as they drive consumers to develop pure, reciprocal and competitive forms of altruism, which, in turn, motivate them to create content.

This study examines the antecedents and consequences of altruistic purpose in the context of gamified review platforms by proposing research questions aimed at eliciting the effects of achievement, social and immersion affordances on altruism, and by providing the first empirical evidence for these paths.

This study provides practical guidance on how review platforms can implement social and immersion affordances to foster altruism and, ultimately, promote user-generated content in the form of comments, photos and videos.

To the best of the authors’ knowledge, the current study is the first to develop a model to predict whether gamification affordances promote forms of altruism that result in user-generated content. The findings will improve practitioners’ strategies by focussing on social and immersion motivational affordances.

Drawing on the pleasure-arousal-dominance (PAD) emotion model, the emotional states of consumers embedded in online reviews can be described through three dimensions, that is, pleasure, arousal and dominance, rather than only the one-dimensional positive and negative polarity, as in previous studies. Therefore, this study aims to explore the effect of online review emotion on perceived review helpfulness based on these three basic emotional dimensions.

A lexicon-based method is developed to analyze PAD emotions of online reviews from JD.com. The zero-inflated negative binomial regression is utilized to empirically validate the study hypothesis. The authors examine the influence of pleasure, arousal, dominance, emotion diversity and emotion deviation on review helpfulness, as well as the moderating effect of product type on the relationship between all independent variables and online review helpfulness.

The study results show that the pleasure emotion impairs the helpfulness of online reviews, while the arousal and dominance emotions have a positive impact. Moreover, the authors find that compared with search products, the effects of pleasure, arousal and dominance on perceived helpfulness are strengthened for experience products. However, the emotional diversity and emotional deviation have opposite effects on the helpfulness of search products and experience products. Additionally, the results show that dominance emotion plays a more important role in the interaction effect.

The empirical findings confirm the applicability of PAD in the online review context and extend the existing knowledge of the influence of review emotion on helpfulness. A feasible scheme for extracting PAD variables from Chinese text is developed. The study findings also have significant implications for reviewers, merchants and platform managers of e-commerce websites.

Aficionados of wine festivals, a component of wine tourism experience, engage in vigorous online discussions that influence fellow travelers’ purchase behaviors. This study aims to delve into these overlooked discussions, identifying emotions, topics and assessing their usefulness in TripAdvisor’s Travel Forums for two US wine festivals: Taste of Yountville and Epcot International Food and Wine Festival, located in traditional and nontraditional wine tourism destinations.

The study uses state-of-art sentiment analysis and topic modeling methods to extract emotions and underlying latent topics in travel forum discussions. Drawing from information theory, two regression analyses are performed on 10,677 forum posts to examine how the extracted Ekman’s emotions and key underlying topics influence the helpfulness of wine forum posts for each festival.

While three topics were identified in Epcot and four in Yountville, both festival platforms highlight travelers’ common preferences for “culinary experience” and “planning” attributes but reveal notable differences in their utility. Other shared novel findings include the importance of “anger” and “surprise” emotions on the helpfulness of forum posts.

These findings enhance wine festival managers’ and destination planners’ understanding of online travelers’ preferences and cognitive evaluation of user-generated contents’ usefulness. This marketing intelligence informs strategies for boosting the wine destination’s economic development.

This research offers a novel comparative analysis of social media on wine festival tourism experiences in diverse regions. Unlike hotel reviews, typically posted after consumption, forums offer unique and broader perspectives on discussions before, during, and after experiencing the wine festival.

This paper aims to explore the possible forms and characteristics of an artificial intelligence (AI) leader and discuss the potential applications of AI in political leadership and governance.

A categorization system consisting of three categories – the level of responsibility, the voting system and the bindingness of the AI’s decisions – was developed to better understand the various types of AI leaders. Additionally, to identify the main characteristics of an AI leader, a comprehensive literature review was conducted. The themes from the literature were then categorized and supplemented with additional discussions.

This paper identifies several potential AI leaders, including the AI President, the AI Dictator, the AI Minister and the AI Consultant. The key characteristics of an AI leader were also discussed. The primary strengths of AI lie in their intelligence and rationality, which could potentially lead our societies toward a peaceful and prosperous future. However, a significant drawback of AI is that it will always be limited by the capabilities and intentions of its programmer, whether human or AI.

Understanding the forms and characteristics of AI leaders may help policymakers and decision-makers explore the possibilities of integrating AI into political leadership and governance.

This paper contributes to the emerging field of AI in governance by exploring the forms and characteristics of AI leaders and discussing their potential applications in political leadership.