This study aims to explore thematic influences on theme park visitors' satisfaction through user‐generated data. To this end, we first used an unsupervised machine learning method, structural topic modeling, and analyzed 112,000 reviews post by visitors to Shanghai Disney Resort from June 16, 2016 to March 4, 2022. Our findings are of great significance for reflecting consumer behavior through user‐generated data. Specifically, we find that visitors' satisfaction is highly related to service in the theme park and their playing feeling, and early tourists pay more attention to the experience of specific playing items while later tourists focus on the overall playing experience. In addition, an empirical study is conducted by treating the ratings associated with each review as dependent variable and each topic represented by comments as independent variables, which shows that the relationship between the customer reviews and ratings by tourists becomes less pronounced over time. In other words, as time goes, customers review can reflect their subjective feelings or experience, but the rating is not. We discover the “dynamics” of user‐generated data over time and gain a better understanding of the aspects and concerns of visitors' satisfaction over time. The findings of the study contribute to the literature on tourism, service, and consumer behavior while also providing valuable practical implications.

Promoting the growth of the lithium battery sector has been a critical aspect of China’s energy policy in terms of achieving carbon neutrality. However, despite significant support on research and development (R&D) investments that have resulted in increasing size, the sector seems to be falling behind in technological areas. To guide future policies and understand proper ways of promoting R&D efficiency, we looked into the lithium battery industry of China. Specifically, data envelopment analysis (DEA) was used as the primary approach based on evidence from 22 listed lithium battery enterprises. The performance of the five leading players was compared with that of the industry as a whole. Results revealed little indication of a meaningful improvement in R&D efficiency throughout our sample from 2010 to 2019. However, during this period, a significant increase in R&D expenditure was witnessed. This finding was supported, as the results showed that the average technical efficiency of the 22 enterprises was 0.442, whereas the average pure technical efficiency was at 0.503, thus suggesting that they were suffering from decreasing returns to scale (DRS). In contrast, the performance of the five leading players seemed superior because their average efficiency scores were higher than the industry’s average. Moreover, they were experiencing increasing scale efficiency (IRS). We draw on these findings to suggest to policymakers that supporting technologically intensive sectors should be more than simply increasing investment scale; rather, it should also encompass assisting businesses in developing efficient managerial processes for R&D.

In Large-Scale Group Decision-Making (LSGDM), effectively implementing consensus models is pivotal for managing decision complexity. While trust-based LSGDM has garnered attention, there remains a need for deeper insights into the dynamics of interexpert trust and the impact of authority effects on the decision-making process. This study introduces a sophisticated model for large-scale group decision-making, incorporating considerations of expert “trustworthiness-authority.” Initially, the study assesses the trustworthiness of experts based on social network relationships and opinion similarity while using background information and consensus levels to establish their authority. Subsequently, experts are categorized into four distinct regions based on their trustworthiness and authority assessments. Furthermore, tailored consensus adjustment methods are proposed for each region based on social contagion theory to facilitate consensus achievement. Additionally, a case study is conducted to demonstrate the rationality and effectiveness of the proposed LSGDM model, considering expert “trustworthiness-authority.” Finally, the necessity and superiority of the proposed model are further verified through comparison analysis and sensitivity analysis.

The aim of this study was to explore the causes of tourists’ lesser enjoyment of theme parks through an unsupervised machine learning approach—structural topic modelling. Specifically, the research extracted a comprehensive list of discussion topics about the travel experience of tourists through the analysis of 112,000 customer reviews released by visitors to the Shanghai Disney Resort from 16 June 2016 to 4 March 2022. Then, we used sentiment analysis to distinguish positive and negative topics and to explore the relationship between tourists who buy different ticket types and sentiments in negative topics. The results show that problems such as “parking,” “service attitude,” “recommendation feeling,” “experience comparison,” and “entrance” may be the main reasons for an unpleasant experience. In addition, we also found that when tourists travel in groups (e.g., via family tickets), customers feel unhappy because of parking and fast track problems. Moreover, when tourists travel alone or with small groups, staff service attitudes, experience comparisons, and entrance processes are the sources of greater concern. Our findings can help theme park managers to better understand the expectations of tourists and take effective measures to tackle issues causing customer dissatisfaction, and they can also contribute to theme park studies in tourism management.

Purpose: This study investigates the evolution of city branding strategies in the digital era, focusing on how aesthetic design elements in city marketing content shape user interactions on social media. It examines how visual design attributes in city imagery drive engagement and uncovers the psychological processes underlying these effects. Design/methodology/approach: Drawing on cue utilization theory and emotion as social information (EASI) theory, the research employs multimodal machine learning to analyze a comprehensive dataset from city marketing organizations. The study utilizes computer vision algorithms and natural language processing to deconstruct visual complexity into three dimensions: feature complexity, compositional complexity and color variety. A negative-binomial regression model is applied to examine the relationships between visual complexity, emotional responses and user engagement. Findings: The findings reveal that feature complexity and color variety exhibit an inverted U-shaped relationship with user engagement, while compositional complexity demonstrates a U-shaped relationship. Additionally, user emotional responses significantly mediate the impact of feature complexity, compositional complexity and color variety on user engagement. These results highlight the nuanced effects of visual complexity on user behavior and the critical role of emotional responses in this process. Originality/value: This study expands the knowledge boundaries of visual marketing through three layers of innovation. First, it overcomes the traditional limitation of treating visual complexity as a unidimensional construct by proposing a triadic analytical framework of “Feature-Composition-Color” tailored to the characteristics of city marketing. This framework uncovers the asymmetric impact mechanisms of different dimensions on user engagement, offering new insights into the long-standing “complexity paradox” in marketing literature. Second, it expands the applicability of emotions as social information (EASI) theory by empirically validating its explanatory power in visually dominant contexts for the first time. It elucidates how visual complexity stimulates deep emotional resonance via the central route while achieving cognitive efficiency through the peripheral route, thereby providing a novel paradigm for multimodal emotional propagation research. Finally, it systematically introduces cue utilization theory from the domain of commercial decision-making into the context of city marketing. This addresses the practical dilemma faced by city marketing organizations on social media, where narrative depth and dissemination breadth often conflict.