Ideas Spread Inc. (E-Journals)
Not a member yet
1368 research outputs found
Sort by
Analysis of Pathways for Preventing and Mitigating Risks in Generative AI Communication
Full text linkThe rapid development of generative artificial intelligence (AI) technology is profoundly reshaping the underlying logic of information production and distribution[9]. However, while enhancing creative efficiency, it also induces global communication risks such as information distortion, value infiltration, copyright disputes, and the erosion of social trust. This paper, from a communication studies perspective, combines framing theory and algorithmic governance logic to analyze the risk generation mechanism intertwined with training dataset bias, the "black box" nature of models, and social network algorithmic distribution. Through comparative research on cutting-edge international governance cases (such as the EU's AI Act) and local Chinese practices, this paper proposes to construct a comprehensive governance framework that is interdisciplinary and multi-dimensional. Research shows that the key to preventing and mitigating risks lies in: establishing a closed-loop algorithmic responsibility system covering the entire process of technology development, content distribution, and user feedback; improving digital watermarking and traceability technology standards for generative content; and deepening international soft law coordination through platforms such as UNESCO. Furthermore, systematically improving public media literacy is the social foundation for preventing the spread of misinformation. The research conclusions of this paper aim to provide theoretical support and practical guidance for building a safe, controllable, and multi-governance communication ecosystem[11]
Spatial Correlation Analysis Between College Students’ Healthy Physical Fitness Adaptability and Airborne Particulates
Full text linkIn order to explore the correlation between PM2.5 and PM10 of outdoor air suspended particulates in university towns in Guiyang City and the performance of college students’ health and fitness adaptation, this research design adopts an ecological study (ecological study) to collect the annual average concentrations of PM2.5 and PM10 in the outdoor air of air quality monitoring stations in colleges and universities from 2022 to 2025. The data on students’ health and fitness adaptability were collected from 67,329 first-year students in various colleges and universities in 2022 and tracked for 4 years to 2025. Cardiorespiratory endurance (800/1000m running), muscle strength, and muscle endurance (1-minute flexion and sit-ups) were further calculated), flexibility (sitting forward bending), explosive power (standing long jump) , and body composition (body mass index), and other indicators of healthy physical fitness of students in colleges and universities . Pearson correlation coefficients were used to analyze the correlation between the concentrations of PM2.5 and PM10 in the outdoor air and various indicators of students’ health and physical fitness. The results of the study showed that the PM10 concentration of suspended particulates in the outdoor air was significantly positively correlated with the students' 800/1000m running seconds. There was a significant negative correlation with the number of 1-minute knee-bent sit-ups for male and female students. There was a significant positive correlation with male and female students’ sitting posture forward bending distance. There was a significant positive correlation with the standing long jump distance of female students. There was a significant positive correlation with female students’ body mass index. In addition, the PM2.5 concentration of suspended particulates in the outdoor air was significantly positively correlated with the students’ 800/1000m running seconds. There was a significant negative correlation with the number of 1-minute knee-bent sit-ups for male and female students. There was a significant positive correlation with male and female students’ sitting posture forward bending distance. There was a significant positive correlation with female students’ body mass index. This study shows that the concentration of suspended particulates in the outdoor air pollutants of university towns in Guiyang is correlated with the test scores of college students’ health and physical fitness
Research on the Application and Ethical Boundaries of Artificial Intelligence Technology in Elderly Care Services
Full text linkAs global population aging accelerates and demographic structures shift, China is facing intensified pressure on its pension security system. While the current diversified elderly care framework—primarily comprising home-based, institutional, and community-based services—has provided some relief, structural challenges such as supply-demand imbalances, a shortage of professional talent, and insufficient institutional safeguards persist. Against this backdrop, the rapid advancement of Artificial Intelligence (AI) offers a transformative technological path for innovating elderly care models. By leveraging technologies such as intelligent monitoring, affective computing, and virtual reality, AI can provide health management, emotional companionship, and social participation, significantly enhancing the efficiency and quality of social services for the elderly.
However, this technology-driven innovation introduces profound ethical considerations. When technological rationality intersects with the core values of social work, risks regarding data privacy, algorithmic bias, and responsibility attribution emerge, potentially widening the digital divide and eroding humanistic care. To address these issues, this study proposes a "human-centered" AI design framework. It systematically explores the application paradigms of AI within the elderly care system and provides a deep analysis of potential value-ethical conflicts. Ultimately, the study attempts to construct a sustainable development path that harmonizes technological innovation with ethical care, providing theoretical references and practical guidance for the standardized evolution of smart elderly care models
The Impact of Clinical Learning Environment on Professional Self-Concept among Undergraduate Nursing Interns: The Mediating and Moderating Role of Proactive Personality
Full text linkTo explore the mediating and moderating roles of proactive personality in the clinical learning environment and professional self - concept of nursing undergraduate interns. Convenience sampling was employed to survey 230 undergraduate nursing students from two undergraduate universities in Henan Province. The correlation analysis results indicated a positive correlation between the clinical learning environment and professional self - concept (r = 0.66, P < 0.01), a positive correlation between the clinical learning environment and proactive personality (r = 0.45, P < 0.01), and a positive correlation between proactive personality and professional self - concept (r = 0.72, P < 0.01). Proactive personality plays mediating and moderating roles in the clinical learning environment and professional self - concept of undergraduate nursing interns. This suggests that nursing educators and managers can enhance the students' professional self - concept by improving the clinical learning environment and cultivating proactive personality
Cross-Task Feature Interaction and Adaptive Intrinsics Learning for Self-Supervised Monocular Depth Estimation
Full text linkSelf-supervised monocular depth estimation is vital for three-dimensional perception in autonomous driving and robotics. Despite significant progress, current methods face limitations regarding isolated feature interaction between depth and pose networks, and the rigid assumption of camera intrinsics. This paper proposes a novel framework with two core improvements. First, we introduce the Mix Query Attention module, which enables deep interaction by using PoseNet-derived motion features as Keys and Values to guide the DepthNet decoder features. This cross-task interaction captures complex geometric and semantic priors, enhancing structural clarity. Second, we propose the Learnable Camera Intrinsics module. Unlike traditional methods that treat intrinsics as fixed, this module parameterizes focal length and principal points as differentiable variables optimized during training, mitigating calibration errors from hardware aging or manual inaccuracies. Evaluated on the KITTI dataset using the Eigen split on an NVIDIA RTX 3090, our method achieves state-of-the-art performance with an AbsRel of 0.071, significantly outperforming Monodepth2, SQLdepth, Manydepth, and SPIdepth. Ablation studies confirm the effectiveness of synergizing feature interaction with adaptive intrinsics calibration for high-fidelity scene reconstruction
Governing Multimodal Consumer Analytics: A Governance-by-Design Reference Architecture for Responsible Technology
Full text linkMultimodal consumer analytics is increasingly deployed in everyday commercial environments, integrating visual, behavioral, transactional, and biometric data to generate continuous behavioral inference. While such systems expand analytical capability, they also intensify governance challenges related to opacity, inferential overreach, accountability displacement, and the normalization of surveillance. Existing research has largely focused on improving model performance and integration techniques, often treating governance as a downstream compliance concern rather than a constitutive design condition. This paper argues that multimodal consumer analytics should be understood as a socio-technical infrastructure whose sustainability depends on governance-by-design rather than unrestricted inferential expansion. To address this challenge, the study proposes the Multilevel Multimodal Data Integration and Analysis Framework (MMDIAF) as a governance-oriented reference architecture for responsible technology. The framework conceptualizes multimodal analytics across interconnected layers of data acquisition, representation management, cross-modal fusion, and decision deployment, while embedding accountability, interpretability, privacy restraint, and contestability as transversal architectural requirements. Rather than introducing a new algorithm, benchmark, or empirical performance evaluation, this paper offers a conceptual and design-oriented contribution, the paper offers a conceptual and design-oriented contribution that translates responsible technology principles into actionable governance intervention points. The framework is further illustrated through an analytical mapping of multimodal consumer analytics in smart retail environments, demonstrating how governance risks emerge through deployment drift and institutional incentives even in systems developed with responsible intent. By reframing multimodal consumer analytics as an institutional governance challenge, this study contributes to responsible technology scholarship by bridging the gap between ethical aspiration and system design. It provides researchers, system designers, and policymakers with a structured reference architecture for maintaining accountability and legitimacy as multimodal inference becomes increasingly embedded in everyday consumer life
Environmental Protection Tax and Green Transformation of Heavy Polluting Enterprises: A DID Empirical Study
Full text linkWith the continuous advancement of green development strategies, environmental regulation has gradually shifted from administrative control to market-based governance mechanisms. The environmental protection tax, as an important institutional arrangement in China’s environmental regulatory system, internalizes pollution costs through taxation and significantly influences corporate production decisions and resource allocation. This study takes the implementation of China’s Environmental Protection Tax Law in 2018 as a quasi-natural experiment and uses data from Chinese A-share listed companies from 2015 to 2022. A difference-in-differences (DID) model is constructed to examine the impact of the environmental protection tax on the green transformation of heavy polluting enterprises. The empirical results show that the environmental protection tax significantly promotes the green transformation of heavily polluting firms. Further analysis indicates that the policy encourages enterprises to increase environmental investment and adopt cleaner production technologies, thereby improving environmental performance and innovation capability. Robustness tests confirm the reliability of the findings. The results provide empirical evidence for improving green taxation systems and offer policy insights for promoting coordinated development between economic growth and environmental protection
Competing Paths to Development: The China Model in Kenya through a Modernization Theory Lens
Full text linkThe rise of China as a major development actor in Africa presents a significant challenge to traditional Western-dominated paradigms. This article uses Modernization Theory as an analytical framework to examine China’s engagement in Kenya, primarily through infrastructure projects under the Belt and Road Initiative (BRI) and adopts a qualitative research design anchored in interpretive analysis. The paper investigates the central paradox of this engagement, noting that while China’s model vigorously promotes the economic drivers of modernization, it simultaneously rejects the socio-political Westernization that Modernization Theory posits as an inevitable consequence. The article contends that Kenya is not a passive recipient but an active agent in this process, navigating a negotiated path of strategic pragmatism and hybridization. By leveraging Chinese financing for tangible projects and maintaining partnerships with Western institutions, Kenya crafts a unique developmental synthesis
From Phased Integration to Sustained Provision: A Study on Mechanism Development for Aesthetic Education Practices at Zhejiang Future Rural College
Full text linkAddressing the structural contradictions in rural aesthetic education practices within the context of the rural revitalisation strategy, this paper focuses on the core research question: ‘How does phased integration occur, and how does it transform into sustained provision?’ The study examines the organisational forms and operational processes of aesthetic education practices at the Zhejiang Rural Revitalisation Academy, covering key aspects such as multi-stakeholder collaboration between government, academia, enterprises, and villages; curriculum and teaching research provision; platform and positional implementation; resource consolidation; and quality review. Building upon existing research that shifts from ‘experience catalogues’ to ‘mechanism explanations,’ this paper proposes and substantiates five institutional pathways: ‘collaborative institutionalisation—curriculum clusterisation—platform operationalisation—resource consolidation standardisation—evaluation verification.’ It reveals that the conditions for generating sustainable provision lie in: stabilising collaborative boundaries through charters and responsibility lists; enhancing supply reproduction capacity via modular curriculum clusters and closed-loop teaching research; cross-cycle operations through platform-base and post systems, educational sedimentation via checklist and archival frameworks, and annual iteration through ‘outcome-process-generation’ three-dimensional indicators and evidence chain verification. Findings indicate that the ‘sustainability’ of Zhejiang Rural Revitalisation Academy's aesthetic education practice stems not from single-project intensity, but from whether the mechanism chain can establish a public supply capacity that is transferable, accumulative, and reviewable. The theoretical value of this paper lies in situating rural aesthetic education within a unified analytical framework of ‘public service provision—grassroots governance capacity—educational quality assurance,’ thereby filling a gap in previous research concerning cross-cycle mechanisms. Its practical value provides a set of implementable, replicable, and verifiable mechanism components alongside evidence-based evaluation interfaces, offering operational pathways for the bounded extrapolation of Zhejiang's experience and the optimisation of local policies
Reconstructing the English Translation Network of the Tao Te Ching: An SNA-Based Study of Indirect Translation
Full text linkThis study investigates indirect translation in the English history of the Tao Te Ching through Social Network Analysis (SNA). Based on a self-constructed corpus of 254 English translations, of which 40 exhibit clear indirect translation practices, translators are modeled as nodes and source-text dependencies as directed ties inferred from paratextual evidence. The network analysis reveals a pronounced core—periphery structure in which a small number of translations function as central hubs, while academically authoritative translations remain relatively peripheral. Drawing on Bourdieu’s field theory as an interpretive aid, the study argues that indirect translation operates as a relational and socially mediated process shaped by translator profiles, publishing dynamics, and collective interpretive patterns