This study constructed and empirically tested an artificial intelligence-generated content-driven whole-process knowledge management framework for the cross-cultural communication of cultural symbols. It was conducted to resolve systemic failures in traditional cultural knowledge transfer, such as low operational efficiency, persistent cultural misinterpretations, and deficient audience co-creation mechanisms. Methodologically, a comprehensive six-month controlled experiment tracked three Chinese cultural symbols across North America, Europe, and Southeast Asia, evaluating the framework against traditional models. Empirical results show the framework cut content production cycles from 24.3 to 3.8 days, reduced labor hours by 91.5%, lowered risk-related costs by 90.4%, and boosted content co-creation users by 900%. Thus, artificial intelligence-generated content serves as a powerful strategic engine, empowering cultural institutions to regain active control of digital knowledge management and foster global value co-creation.
This study developed a distortion risk function and an echo chamber intensity model to analyze the impact of generative artificial intelligence (AI) on information risks in higher education. This was done because AI lowers fabrication costs and maximizes realism, creating a structural crisis that traditional manual review paradigms cannot manage. To investigate, researchers captured social media data from an AI-forged campus event and a traditional rumor event, using an Susceptible-Infected-Recovered (SIR) epidemic model to simulate propagation dynamics and intervention windows. The study found that AI-generated rumors had a basic reproduction number 137% higher than that of traditional ones, compressing the peak arrival time to 6.2 hours. The critical intervention window shrank to under two hours, and cross-disciplinary propagation surged. This means traditional linear response mechanisms have systematically failed. Universities must adopt a collaborative governance framework utilizing digital watermarks, algorithmic literacy education, and dynamic ethical regulations to manage future risks.
This study explored the nonlinear link between digital transformation and audit quality in manufacturing alongside supply chain concentration effects. It was conducted because traditional linear research ignores complex digitalization risks, like algorithmic black boxes, failing to fit manufacturing’s unique financial integration contexts. The authors applied a framework combining mathematical deduction of misstatement risks, game theory, and panel regression on data from 12,450 Chinese A-share manufacturers from 2012 to 2022. The findings revealed an inverted U-shaped relationship where early transformation raises risks and lowers audit quality up to a critical point (Digital Transformation (DT)=1.58)—after which standardized governance reduces risks. Additionally, high supply chain concentration accelerates this benefit. This means managers must strategically pace information technology investments around this threshold, while auditors should adapt resource allocation to specific digital maturity stages to safely capture transparency dividends.
This study proposes a four-in-one AI-assisted model for packaging design and empirically evaluates it using 120 comparative projects. Traditional design models based on experience, templates, functionality, or imitation produce homogenized designs, long development cycles, high costs, and limited ability to meet contemporary demands for personalization, sustainability, and e-commerce. Compatibility analysis mapped AI capabilities to design needs, resulting in a model integrating data-informed generation, human–AI optimization, scalable tools, and blockchain governance. Stratified sampling and regression analyses showed that AI reduced development cycle times by up to 68.3%, doubled creative concepts, increased satisfaction by 1–1.5 points, improved adoption by 12–15 points, halved revisions, reduced costs by 36.9%, and increased ROI by 88%, with SMEs benefiting the most. The findings demonstrate AI’s scalable value in packaging design through human collaboration, particularly for smaller firms, highlighting the need for improved cultural modeling, sustainability tools, and IP frameworks to support industry transformation.
This paper develops a composite reinforcement learning framework for energy industry supply chains to systematically analyze state transitions and control pathways. The research is motivated by the vulnerability of energy networks to market instability and operational disruptions, which reduce the effectiveness of static optimization models because of inadequate information propagation. The proposed framework employs a hierarchical parallel architecture that integrates real-time feedback loops and data quality metrics into the reward mechanism. Using enterprise operational log data, empirical evaluations simulated scenarios involving extreme weather and data interruptions. The findings show that mean scheduling efficiency increased from 55.3 to 65.8, while the system demonstrated a substantial improvement in recovery speed under extreme disturbances. These results confirm that integrating algorithmic optimization with human-in-the-loop information governance significantly enhances organizational resilience and system self-healing capabilities.
This study developed a dynamic financial health evaluation model that integrated multi-source incomplete views and exponential time decay weights. The framework addressed critical bottlenecks in traditional assessments, including decision lag from static statements, systematic bias caused by missing heterogeneous data, and “concept drift” where old samples failed to reflect current market cycles. The model fused financial ratios, macro indicators, and Natural Language Processing-extracted sentiment. It employed variational inference with masking to reconstruct missing features, Least Absolute Shrinkage and Selection Operator for dimensionality reduction, and a gradient boosting decision tree adjusted for temporal relevance. Empirical tests on A-share manufacturing data yielded 91.2% accuracy and an Area Under Curve (AUC) of 0.935. Notably, the model maintained high robustness (0.852 AUC) despite a 40% missing data rate. These results signified a paradigm shift toward proactive risk management, offering a rigorous tool for early risk intervention.
In this study, the author developed a three-dimensional cognitive, emotional, and behavioral resonance model for managing information ecosystems in algorithm-mediated organizations. She addressed the structural suppression of mission-critical content by entertainment-centric algorithms that prioritize short-term engagement over long-term organizational value. Using system dynamics and 400,000 data logs from the Douyin application programming interface, she employed propensity score matching and multi-objective reinforcement learning (Soft Actor-Critic) to optimize content weightings, followed by a 30-day A/B test on Douyin. Results indicated that the model significantly increased click-through rates by 29.4%. This finding suggests that integrating affective markers into algorithmic design effectively aligns technological efficiency with strategic organizational responsibility, ensuring that vital discourse survives the entertainment-driven digital deluge.
This study presented an intelligent visual design assistant system to address limitations in traditional tools that focused on template retrieval and lacked personalized recommendations. Modern design requires diversified, high-frequency output, but current research often ignores the multi-link collaborative nature of the process. Using a convolutional neural network and transformer architectures, five modules, including semantic coding and layout recognition, were used to extract features and analyze requirements. A dual closed-loop mechanism iteratively optimized solutions via user feedback and designer behavior. Results from control experiments showed that style recognition accuracy rose to 89.4%, modification rounds fell from 4.6 to 2.8, and task time decreased by 28.3 minutes. The system achieved a balance between efficiency and quality. It extended deep learning from single-task evaluation to full-process design, supporting industry digital transformation.
This study constructed and empirically validated a blockchain-enabled multiloop collaborative model for managing intangible-cultural-heritage information resources. It addressed structural dilemmas in intangible-cultural-heritage digitalization, including ownership ambiguity, centralized storage risks, and insufficient participant incentives. The methodology involved designing a system for on-chain registration and smart contract-driven incentives, which was tested using large-sample operational data from four mainstream Chinese digital platforms. Results indicated a 98.1% success rate for the primary traceability process and confirmed that dynamic incentives significantly boosted node activity; however, high process rollback ratios and manual intervention costs were identified as primary operational constraints. These findings established a theoretical and practical framework for shifting from static archiving to dynamic, trustworthy governance, offering a replicable path for optimizing organizational efficiency and ensuring the sustainable activation of cultural assets in the digital era.
One possible explanation for algorithm aversion is that people’s general understanding of algorithms is poor. According to this logic, greater algorithm literacy would be expected to reduce such aversion. This paper argues, however, that algorithm literacy might generate unintended consequences by making the flaws and limitations of algorithms more salient to users. Using a laboratory experiment, this study examines the relationship between algorithm literacy and algorithm aversion in medical decision making. Contrary to what many people might assume, higher algorithm literacy was associated with greater algorithm aversion. The study's findings extend research on algorithm aversion by showing that algorithm literacy may amplify rather than reduce algorithm aversion, calling into question its role as a mitigation strategy.
This study constructs a big data-driven process optimization model for a financial sharing center. It addresses the research gap regarding the empirical analysis of real process log data in integrating big data capabilities with organizational efficiency. The methodology involves analyzing 24 months of event logs from a manufacturing group using process mining, state machine modeling, and fixed-effect regression to evaluate operational efficiency. Results indicate that big data empowerment significantly shortens node waiting times, reducing rule pre-check delays by 32.5%, and improves first-pass yields. Findings also reveal that process complexity and analytics intensity influence efficiency gains, with high-frequency manual processes showing the greatest improvement. These results imply that a data-driven closed-loop system is essential for enhancing organizational effectiveness, intelligent financial management, and sustainable digital transformation.
This study constructed and empirically validated a blockchain-enabled multiloop collaborative model for managing intangible-cultural-heritage information resources. It addressed structural dilemmas in intangible-cultural-heritage digitalization, including ownership ambiguity, centralized storage risks, and insufficient participant incentives. The methodology involved designing a system for on-chain registration and smart contract-driven incentives, which was tested using large-sample operational data from four mainstream Chinese digital platforms. Results indicated a 98.1% success rate for the primary traceability process and confirmed that dynamic incentives significantly boosted node activity; however, high process rollback ratios and manual intervention costs were identified as primary operational constraints. These findings established a theoretical and practical framework for shifting from static archiving to dynamic, trustworthy governance, offering a replicable path for optimizing organizational efficiency and ensuring the sustainable activation of cultural assets in the digital era.
In this study, the author investigates the impact of government subsidies on the digital transformation of manufacturing enterprises, focusing on the moderating role of market competition. She addresses the "wait-and-see" dilemma caused by high digital investment costs and fills the research gap regarding how market structures interact with financial incentives. She analyzed panel data from 7,328 Chinese A-share manufacturing firms (2010-2024) using fixed effects, two-stage least squares, and panel threshold regression, with digital levels measured via term frequency-inverse document frequency text mining. Results revealed that subsidies significantly catalyzed digitization, but the marginal benefit followed an inverted U-shaped relationship with competition. The subsidy effect peaked at 0.46 when competition was moderate (Herfindahl-Hirschman Index of 0.28-0.46); extreme competition diminished this policy efficiency significantly. High-quality accounting information amplified this effect by 12%. These findings imply that policy must be differentiated based on "competition windows" and that firms should improve transparency to maximize the digital dividend.
Effectively managing data as a core information resource is critical as the digital economy shifts from demographic to data dividends. While existing literature emphasizes artificial intelligence's (AI's) substitution effect, it overlooks how AI activates data factors. Positioning AI as a key tool for information resource management, this paper constructs an integrated framework of micro-level factor activation, meso-level friction reduction, and macro-level spatial reconstruction. Using panel data from 285 Chinese cities (2013-2023) and spatial Durbin model with dynamic effects, the following was found: (1) AI nonlinearly enhances data factors' output elasticity, confirming its activation effect on information resources; (2) a real economy threshold (S & lowast; = 0.342) exists-below it, AI cannot reduce industrial information friction; and (3) virtual agglomeration substitutes for geographic distance, revealing digital gravity's role in reshaping regional economies. Three differentiated models-factor activation, friction reduction, and cloud collaboration-are proposed to help regions maximize value from their information resources.
This study developed a big-data-driven sports injury risk prediction model and preventive training management framework to address the limitations of traditional, experience-based injury assessments. It aimed to improve early detection and intervention effectiveness amid the growing complexity of multisource training data. Using wearable devices, questionnaires, and medical records from 210 athletes, multidimensional variables such as training load, fatigue, sleep, and historical injuries were integrated through an eXtreme Gradient Boosting–long short-term memory hybrid model within a closed-loop feedback system. The model achieved an area under the curve of 0.814 and an F1 score of 0.689, correctly identifying 72% of high-risk athletes. Following stratified interventions guided by model outputs, injury rates decreased from 11.2% to 6.1%, while training compliance improved notably. These results demonstrated that intelligent, continuously adaptive predictive modeling could enhance sports health management by transforming fragmented monitoring data into actionable prevention strategies and individualized feedback loops.
In the digital era, traditional and high-art dance forms often struggle to gain visibility on social media due to their structural complexity, which limits user imitation and algorithmic promotion. This study addressed this challenge from an information resource management perspective by developing a big data-driven model of communication potential. Using empirical data from TikTok, it was demonstrated that strategically simplifying movement complexity, such as through key-frame extraction, significantly enhances user engagement and network diffusion while preserving core cultural meaning. The findings provide organizations with a practical, data-informed approach to managing intangible cultural assets in algorithm-mediated environments, balancing cultural integrity with platform-driven dissemination logic.
This study presents an artificial intelligence-driven closed-loop system for college English translation teaching, transforming student learning activities into structured information resources for institutional decision-making. By integrating artificial intelligence scoring, multimodal data, and human-in-the-loop evaluation, the system enables dynamic monitoring, anomaly detection, and process-level optimization. Empirical results demonstrate improvements in translation accuracy, cultural adaptability, and feedback efficiency, while revealing operational challenges under high workload. From an information resources management perspective, this framework illustrates how educational processes can be leveraged as strategic organizational assets, supporting adaptive management, resource allocation, and sustainable decision-making.
An empirical framework for the spatio-environmental transformation of industrial heritage parks was proposed to address urban renewal challenges, such as single spatial structures, unsustainable environments, and inefficient information management. Based on information resource management theory, a multi-source information integration approach was adopted, integrating remote sensing, behavior logs, policy documents, and social survey data into a space–environment–society–policy dynamic coupling model. Findings show that multidimensional social feedback and policy adjustment increased these variables’ weight by 23.6%, mitigating information heterogeneity and quality instability. Results also highlight spatial heterogeneity, with environmental gains varying across units and not always synchronizing with social satisfaction. This study demonstrates that heritage transformation is a dynamic, iterative process, providing a practical data-driven path for synergizing spatial renewal, ecological health, and community needs, as well as a verifiable information resource management framework.
Ship electronic chart display and information systems have become critical information resources in ocean navigation; however, their effectiveness has been frequently undermined by data overload, delayed responses, and poor interface design. This study treated ship electronic chart display and information systems as a managed information resource needing to be aligned with operational demands and human cognitive capacity—not just a technical tool. The study proposed a lightweight, three-layer optimization approach based on adaptive data loading, predictive risk awareness, and cognitive load-aware display. This approach improved decision support without hardware upgrades. Validated through real automatic identification systems (AIS) trajectories and crew eye-tracking, the framework enhanced situational awareness and reduced operator stress in high-traffic voyages. The findings offer practical guidance for maritime organizations regarding better management of onboard information systems to improve safety and operational efficiency.
This study tested how intelligent writing aids change literary creation outcomes and workflows. It was done to determine whether artificial intelligence (AI) boosts efficiency and innovation without eroding style, diversity, and author control. The authors ran a three-group experiment comparing human-AI co-writing, AI-only writing, and manual writing using 135 works across varied prompts, capturing platform logs and blind ratings for innovation, diversity, satisfaction, revision rounds, and text confusion. Co-writing produced the highest innovation, diversity, and satisfaction and the most revision loops, but it also raised confusion and occasionally triggered structural templates, theme drift, and stylistic breaks; satisfaction peaked at moderate confusion and declined when iteration became excessive. This means AI tools work best as human-led collaborators: they can expand options and reduce writer’s block, but creators need to exercise governance in the way of prompt structure discipline, revision limits, and style checks to protect originality and maintain a coherent voice.
