Yang Cheng

Purpose Ethics enables organizations to effectively resolve dilemmas while acting socially responsible. This study aims to examine how current communication practices involving AI technologies align with domain-specific Generative AI (GenAI) guidelines to foster the quality of organizational-public relationships (OPR). Design/methodology/approach The discussion on ethical principles governing AI in computer science, along with the conceptualization of OPR linked to professional codes of ethics, informed by the feminist philosophy known as the ethics of care, contributes to the development of a proposed AI ethics framework to foster OPR. Drawing upon this framework, the use of content analysis to unpack industry discourse reveals the extent to which the industry’s understanding of GenAI ethics aligns with this proposed framework. The implications of implementing this framework to foster OPR are discussed. Findings Communication professionals view social responsibility and authenticity crucial for ensuring ethical AI practice, with truthfulness, respect and equity following closely. However, adherence to ethical AI use in communication depends on the implementation of explainability, accuracy, fairness and machine autonomy in computer science. Embracing the ethics of care to integrate these ethical principles into the current AI ethics framework in communication becomes crucial for easing this tension. Originality/value The proposed AI ethics framework bridges AI ethics between communication and computer science by capturing social responsibility, authenticity, truthfulness, respect and equity. This framework helps shape professional codes of ethics to address challenges in the rapidly evolving AI-driven communication landscape and advocates for the engineering of responsible AI tools to foster quality OPR. The outcomes advance cross-disciplinary research and cross-sectoral practices of AI ethics by leveraging the ethics of care, thereby connecting AI ethics across computer science and other fields.

Overconsumption, which leads to considerable adverse effects on sustainability, is a significant concern in the tourism sector, especially under flat-rate pricing (e.g., buffets), where consumption is unlimited after consumers pay a fixed price. When charged with higher prices, will people consume more regardless of how much they need? Previous findings are mixed. To answer this question, by conducting three laboratory experiments and one field experiment, our research introduces an influential ideology and shows that this higher-price-driven overconsumption behavior only holds among people who believe in the Protestant Work Ethic, an influential concept used to explain the historical rise of capitalism. People who believe in the Protestant Work Ethic emphasize input-outcome correspondence and are therefore more likely to feel entitled to consume more as their expenditure increases. This belief leads them to overconsume under flat-rate pricing. In contrast, people who do not subscribe to the Protestant Work Ethic consume similar quantities regardless of their spending. Our findings extend the Protestant Work Ethic literature, revealing how this virtuous construct can lead to not-so-virtuous behavior like overconsumption. Marketers can leverage this to influence consumer overconsumption, enhancing profitability and sustainability.

We investigated whether a pre-change representation is inhibited or weakened under correct change detection. Two arrays of six objects were rapidly presented for change detection in three experiments. After detection, the perceptual identification of degraded stimuli was tested in Experiments 1 and 2. The weakening of a pre-change representation was not observed under correct detection. The repetition priming effect was observed for a pre-change object and the magnitude was equivalent to the effect for a post-change object. Under change blindness, repetition priming for a pre-change representation was observed when detection did not require report of location in Experiment 1 and was not observed when location was required to be reported in Experiment 2. The results of Experiment 3 showed that a pre-change representation was recognized at a higher rate under correct detection than under change blindness, reflecting a stronger rather than a weaker pre-change representation in the former context

The global spread of the novel coronavirus is affected by the spread of related misinformation—the so-called COVID-19 Infodemic—that makes populations more vulnerable to the disease through resistance to mitigation efforts. Here, we analyze the prevalence and diffusion of links to low-credibility content about the pandemic across two major social media platforms, Twitter and Facebook. We characterize cross-platform similarities and differences in popular sources, diffusion patterns, influencers, coordination, and automation. Comparing the two platforms, we find divergence among the prevalence of popular low-credibility sources and suspicious videos. A minority of accounts and pages exert a strong influence on each platform. These misinformation “superspreaders” are often associated with the low-credibility sources and tend to be verified by the platforms. On both platforms, there is evidence of coordinated sharing of Infodemic content. The overt nature of this manipulation points to the need for societal-level solutions in addition to mitigation strategies within the platforms. However, we highlight limits imposed by inconsistent data-access policies on our capability to study harmful manipulations of information ecosystems.

As the negative environmental impacts of transportation systems become more severe, governments and environmental groups are seeking more sustainable transportation options, such as replacing fuel-powered vehicles with electric vehicles and expanding public transportation systems to reduce the number of people driving on their own, in order to reduce the environmental impacts of transportation systems. At present, the rapid expansion of public transportation systems is not an easy task and requires a long period of time to plan for expansion and construction, so people are increasingly looking to find means of transportation that meet sustainable conditions as solutions. In this context, electric bicycles are one of the solutions that people can choose, with benefits such as energy saving, carbon reduction, effective air pollution reduction, and simple and labor-saving riding. However, in Taiwan, despite the many benefits of electric bicycles, their popularity is not high. Therefore, this study focuses on the factors that affect the purchase of electric bicycles in Taiwan. The Influential Network Relation Map generated by the Z-based Decision-making Trial and Evaluation Laboratory technique is used to describe the influence relationships among the factors and to establish the key evaluation criteria of electric bicycle purchase intention. The results indicate that vehicle price, safety, motor performance, battery life, and battery durability are the most important factors in purchasing electric bicycles. Furthermore, the power of motor is considered as the factor that most significantly affects other criteria, while safety and price are most likely to be affected by other criteria. This study has contributed to academia and industry, for the dependency weights of these factors are set to provide a scientific and systematic way to show how consumers think in the decision-making process and to provide more reliable information and management implications for the electric bicycle industry.

In this paper, according to the perspective of customers and products, by using clustering analysis and association rule technology, this paper proposes a cross-marketing model based on an improved sequential pattern mining algorithm, where an improved algorithm AP is applied. The algorithm can reduce the time cost of constructing a projection database and the influence of the increase of support on the algorithm efficiency. The improved idea is that when the first partition is used to generate the projection database, the number of itemsets in the projection database is sorted from small to large, and when the second partition is used, the sequence patterns are generated directly from the mined sequence patterns, so as to reduce the construction of the database. The experimental results show that this method can quickly mine the effective information in complex data sets, improve the accuracy and efficiency of data mining, and occupy less memory consumption, which has good theoretical value and application value.