This study focuses on the application mechanism of human-robot collaboration (HRC) in the assembly process of production operations. Taking Zeekr 5G Smart Factory, cosmetic packaging production line, and European item small-batch assembly as cases, based on the ISO/TS 15064 four-level collaboration framework and the "static capability table-dynamic event trigger" model, it adopts multi-case comparison and performance normalization methods to construct an HRC evaluation system for assembly links from three dimensions: task allocation, safety mechanism, and performance improvement. The results show that three modes, namely heavy-load synchronous collaboration, light-load serial collaboration, and light-load hybrid fusion, all significantly improve efficiency and flexibility. Among them, task separation degree and collaboration level have a significant positive impact on efficiency, quality, and flexibility, and different scenarios show significant differences between safety costs and performance benefits. The study verifies the effectiveness of the task-risk-level chain framework, providing theoretical and practical references for manufacturing enterprises to select appropriate collaboration levels, optimize safety investment, and promote the transformation of intelligent manufacturing according to load, beat, and batch characteristics.

Against the backdrop of the long-term coexistence of street-hailing taxis and ride-hailing services, unauthorized surcharging, fare bargaining, and detours that effectively inflate passengers’ payments occur from time to time. Such practices may increase drivers’ revenues in the short run, but they can also trigger complaints, erode service trust, and raise regulatory enforcement costs. This paper abstracts the taxi and ride-hailing sectors as two competing players and develops a static complete-information game-theoretic model that incorporates regulatory penalties and reputational losses, in order to characterize the Nash equilibria under two strategies: compliant pricing and unauthorized surcharging. The analysis shows that whether unauthorized surcharging becomes prevalent in the industry hinges on the gap between the incremental gains from surcharging and the expected costs of punishment. When the additional payoff from surcharging exceeds the sum of the expected penalty loss upon detection and the reputational loss, both parties tend to adopt unauthorized surcharging, yielding the (unauthorized ,unauthorized ) equilibrium; otherwise, the (compliant ,compliant ) equilibrium emerges. Based on the derived threshold condition, the paper proposes policy implications including increasing the probability of detection, strengthening penalty severity, improving credit-based evaluation mechanisms, and enhancing platform–regulator collaborative governance.

The widespread adoption and inherent convenience of third-party payments have significantly reshaped the consumption habits and structure of college students. This kind of social influence, manifested as imitating others and following group norms, is a key factor influencing college students' consumption choices. This study, through the method of literature review, referred to relevant research on this topic both domestically and internationally, clarified the key concepts, examined the impact of such payments on the consumption of higher education students, and explored the effect of social influence in it. This study underscores that in the digital economy, understanding the interaction between payment platforms and social influence is crucial for promoting responsible consumption among college students.

Based on the practical need for financial support to increase farmers' income within the context of rural revitalization, this paper explores the impact of agricultural insurance and agricultural loans on farmers' income and their synergistic mechanism. The study aims to clarify the directional effects, regional disparities, and interactive outcomes of these two factors on sustainable farmers' income, providing empirical evidence and policy references for improving the rural financial service system. Utilizing panel data from 31 provinces in China from 2012 to 2022, the research conducts an empirical analysis using the SYS-GMM model and panel threshold model. The results indicate that both agricultural insurance and agricultural loans exhibit a significant inhibitory effect on the growth of farmers' income, suggesting that the current synergistic mechanism between agricultural insurance and agricultural loans fails to effectively promote income growth and even demonstrates a "synergy failure" phenomenon. Further analysis reveals that this negative impact shows distinct regional heterogeneity, being particularly pronounced in eastern China, while relatively moderate in central and western regions. The conclusion posits that the current development and linkage model of agricultural insurance and agricultural loans in China have not yet fully realized their income-enhancing effects, and their efficacy is influenced by regional economic conditions and development stages.

This review systematically analyzes cutting-edge fuel cell technologies, focusing on proton exchange membrane fuel cells (PEMFCs), solid oxide fuel cells (SOFCs), and emerging metal-air systems. It elaborates on material innovations in catalysts and electrolytes, automation-driven manufacturing processes, and digital twin-based health management strategies. Case studies validate the performance enhancements, while current challenges and future development directions are outlined to provide a comprehensive reference for research and industrial applications.

The impact of cooperative network on enterprise innovation performance has been paid attention by the academic community. The importance evaluation of cooperative network on enterprise key core technology breakthroughs is the focus of this paper. This paper takes China's high-tech manufacturing industry from 2010 to 2020 as a research sample to build a cooperative network, uses a random forest-based machine learning model to measure the im-portance of cooperative network to the key core technology breakthroughs of enterprises, and further analyzes the relative importance of cooperative network features to fit the key core technology breakthroughs of enterprises. The findings are as follows: (1) Overall, the lack of cooperative network features will reduce the fitting performance of key core technology breakthroughs. (2) Rank the features in the cooperative network according to the degree of im-portance, and further use the partial dependence diagram to visualize the marginal impact of the three most important features, individual network size, number of intermediaries and proximity centrality, on the breakthrough of key core technologies of enterprises.

As a core element driving technological progress and cultivating innovative talents, the spirit of scientists holds irreplaceable value in ideological and political education and scientific literacy development for contemporary universities. Through literature analysis and current situation assessment, this study systematically investigates the understanding and dissemination characteristics of the scientific spirit among college students. It identifies existing challenges including monotonous communication formats, insufficient content depth, weak international outreach, and low student engagement. The paper proposes targeted improvement strategies across four dimensions: innovating communication methods, enriching content, enhancing international communication, and boosting student participation. These findings provide valuable references for universities to strengthen the educational function of the scientific spirit and nurture scientific talents capable of undertaking the great mission of national rejuvenation.

In response to the limitations of traditional static and subjective methods, this study pioneers an AI-driven framework for assessing media convergence competitiveness by leveraging natural language processing and machine learning. Analyzing a large-scale dataset of YouTube comments from January to June 2024, the research employs sentiment classification models—including Support Vector Machine, Random Forest, and Gradient Boosting—with Gradient Boosting demonstrating superior performance. The findings reveal that the proposed framework not only outperforms conventional approaches in accuracy and consistency but also captures nuanced public sentiments and their temporal dynamics, uncovering a dual pattern of instant emotional reactions and reflective long-term evaluations. This provides policymakers and media practitioners with a scalable, data-driven tool for real-time assessment and evidence-based strategy formulation. Ultimately, the study underscores the transformative potential of integrating computational social science with media studies, significantly enhancing the precision and predictive power of competitiveness evaluation beyond qualitative paradigms.

Graph neural networks (GNNs) achieve deep representation learning of graph-structured data through a message passing mechanism, becoming a core technical paradigm for processing structured data. This paper systematically analyzes the architectural principles of the message passing mechanism, illustrating its three-step core process of "message generation-aggregation-update" and its mathematical implementation. The paper focuses on comparing the mechanism with typical models such as GCN and GAT. Drawing on empirical examples from fields such as bioinformatics, financial risk control, and retail supply chain, the paper reveals the unique advantages of this mechanism in capturing entity associations and discovering hidden patterns. In protein function prediction, the message passing-based DeepFRI model achieves accurate mapping between structure and function; in financial fraud detection, aggregating multi-hop transaction information significantly improves the ability to identify hidden criminal networks. The paper also analyzes current challenges such as oversmoothing and large-scale graph adaptation, proposing solutions such as residual connection optimization and neighbor sampling. This research demonstrates that the message passing mechanism provides a modeling approach that captures the relational nature of structured data, demonstrating strong potential for application in complex, cross-domain problems. Future applications will be further overcome by combining self-supervised learning with large-scale model technologies.

This study investigates the accessibility of the land transport network in Côte d'Ivoire using ArcGIS software to analyze accessibility across various spatial scales. The findings reveal that the accessibility to most regions in the country is generally favorable, with the majority of areas reachable within 3 hours of travel time. The spatial distribution of accessibility follows a core-periphery pattern, where accessibility time increases as one moves further from administrative centers. Interestingly, higher-level administrative centers tend to have poorer accessibility, primarily due to their fewer numbers and more dispersed locations. The overall accessibility time for various administrative divisions typically remains within 2 hours. When compared to the original administrative boundaries, the variation in the service area based on transportation accessibility typically does not exceed 60%. The analysis also indicates that regions with stronger economic development, denser road networks, and smaller administrative units tend to have larger service areas. This research offers valuable insights into the planning of Côte d'Ivoire’s transport network, providing recommendations for improving regional connectivity and enhancing the efficiency of administrative operations through better transport accessibility.