Open Access

View Full Issue

Addressing Perceived Resistance to Biometric Security Systems in Airports: Exploration With a General Bayesian Network-Based Decision Support System View Full Issue

Ensuring robust security is paramount in the era of global air travel. As airports and other organizations increasingly deploy biometric security systems, end-user resistance poses significant... Show More

Ensuring robust security is paramount in the era of global air travel. As airports and other organizations increasingly deploy biometric security systems, end-user resistance poses significant challenges for managers striving to balance security, efficiency, and acceptance. This study develops an interpretable DSS to manage such resistance, utilizing a GBN derived from 339 passenger surveys regarding airport biometric e-gates. The GBN models how perceived risks, compatibility, and trialability shape three distinct resistance outcomes. Building on this model, the authors conduct simulation-based “what-if” analyses across three intervention scenarios to examine how specific design and policy choices can mitigate resistance. An expert evaluation by airport security managers indicates that the DSS's recommendations are realistic and offer improvements over current heuristic strategies. While live field deployment remains a subject for future research, the proposed framework offers a reusable blueprint for DSS design in other biometric and AI-enabled security applications.

Open Access

View Full Issue

Game-Theoretic Decision Rights Allocation for Cross-Enterprise Data Sharing Under the Federated Learning FATE Framework Under the Data Legal Context: An Automotive Supply Chain Study View Full Issue

In modern automotive supply chains, enterprises such as manufacturers, component suppliers, and logistics providers are tightly interconnected yet reluctant to share operational data due to privacy... Show More

In modern automotive supply chains, enterprises such as manufacturers, component suppliers, and logistics providers are tightly interconnected yet reluctant to share operational data due to privacy, competitive, and regulatory concerns. While federated learning (FL) offers a technical pathway for collaborative model training without exposing raw data, most existing frameworks neglect the governance challenge of allocating decision rights among partners with diverse data quality, volume, and computational resources. This study proposes a game-theoretic decision rights allocation mechanism integrated into the FATE federated learning platform, designed to ensure fairness, efficiency, and stability in cross-enterprise data sharing. The method models each participant's contribution through a payoff function incorporating data utility, timeliness, and cost, and determines decision influence by solving for a cooperative Nash equilibrium under privacy constraints.

Open Access

View Full Issue

Employee Turnover Prediction Research of Human Resource Management on Machine Learning Algorithms and Big Data Analysis View Full Issue

This study introduces a new tool for predicting employee turnover using machine learning (ML) and big data. This method integrates LightGBM and XGBoost (both weighted 1, with predictions summed) to... Show More

This study introduces a new tool for predicting employee turnover using machine learning (ML) and big data. This method integrates LightGBM and XGBoost (both weighted 1, with predictions summed) to enhance accuracy and stability. To improve model interpretability, the SHAPT model is used to identify key factors affecting turnover, such as salary, position, and tenure. Experimental results show the integrated model outperforms standalone LightGBM and XGBoost: accuracy is 1.5% higher, F1 value is 0.02 higher, and AUC reaches 0.9504. These validate the model; SHAP analysis also provides actionable HR management insights, enabling early identification and response to potential employee departures. The research offers practical tools for HR decision-making. Future work will incorporate additional socio-economic variables and dynamic data to further improve prediction performance.

Open Access

View Full Issue

Analysis Theories on Artificial Intelligence, ChatGPT, Data Science, and Metaverse: The Case of Digital Medicine View Full Issue

Healthcare organizations are increasingly adopting digital technologies, with Artificial Intelligence (AI), Data Science, and the metaverse driving significant advancements in smart healthcare. AI... Show More

Healthcare organizations are increasingly adopting digital technologies, with Artificial Intelligence (AI), Data Science, and the metaverse driving significant advancements in smart healthcare. AI facilitates personalized medicine and efficient drug development, while Data Science enables predictive analytics and big data management, enhancing patient outcomes and healthcare quality. The metaverse introduces immersive training and telemedicine platforms, revolutionizing patient engagement and healthcare research. This study conducts a scoping review of 6,171 articles, analyzing the transformational impact of AI, ChatGPT, Data Science, and the metaverse on healthcare. It highlights the benefits and risks of these technologies, identifies research gaps in their application within the healthcare field, and proposes new directions for advancing digital medicine. By exploring these developments, the research provides insights into the continuous evolution of healthcare through innovative digital solutions.

Open Access

View Full Issue

The Relationship Between Leadership Decision-Making Styles and Employee Performance in Government Public Sector: A Multi-Case Comparative Study View Full Issue

Leadership decision-making styles exert a significant influence on employee performance, yet the underlying mechanisms are seldom linear, as heterogeneity, nonlinear responses, and cross-level... Show More

Leadership decision-making styles exert a significant influence on employee performance, yet the underlying mechanisms are seldom linear, as heterogeneity, nonlinear responses, and cross-level dependencies often complicate the relationships. To address these complexities, this study proposes a Hybrid Multi-Method Framework (HMMF) that integrates four complementary perspectives: symmetric structural modeling to estimate direct and mediated paths, configurational analysis to capture equifinality and causal asymmetry, necessary-condition testing to identify noncompensatory constraints, and cross-level evaluation to account for organizational context. Applied to diverse organizational settings, HMMF examines how leadership styles, mediators, and moderators jointly shape performance and is benchmarked against widely used single-paradigm approaches such as PLS-SEM, CB-SEM, and fsQCA. The evaluation covers explanatory power, predictive relevance, configurational strength, and robustness, and results show that HMMF consistently outperforms these baselines.

Open Access

View Full Issue

Innovative Applications of Diffusion Models in Visual Style Transformation for Brand Logos View Full Issue

Brand logos anchor visual identity, yet adapting them to diverse styles is difficult because geometry, typography, and brand cues must be preserved while appearance changes. The authors present... Show More

Brand logos anchor visual identity, yet adapting them to diverse styles is difficult because geometry, typography, and brand cues must be preserved while appearance changes. The authors present LogoDiffusion-Align (LDA), a diffusion framework with three coordinated modules: Structure-Preserving Control (SPC) constrains shapes and text to prevent geometric drift; Style-Consistent Alignment (SCA) injects learned style tokens to achieve coherent, scene-wide stylization; and a Logo-specific Identity Module (LIM) embeds brand-aware representations to retain distinctive identity features. Across multiple datasets and usage scenarios, LDA outperforms strong diffusion-based baselines including ControlNet, DreamBooth, StyleTokenizer, and InST on both fidelity and identity preservation. In controlled comparisons, LDA attains higher SSIM (0.789 vs. 0.742) and CLIP-Id (0.752 vs. 0.708), while also reducing FID and LPIPS, indicating a more favorable fidelity–perceptual quality trade-off.

Open Access

View Full Issue

The Impact of Corporate Social Responsibility Decisions on Brand Value: A Case Study of Alibaba View Full Issue

Corporate social responsibility (CSR) has become a strategic element in the digital economy, especially for platform enterprises where brand value depends heavily on public trust. Most existing... Show More

Corporate social responsibility (CSR) has become a strategic element in the digital economy, especially for platform enterprises where brand value depends heavily on public trust. Most existing studies, however, focus on manufacturing or consumer goods, leaving limited evidence for platforms such as Alibaba. To address this gap, the study employs three methods: a partial least squares structural equation model to test mediation through brand strength, a threshold regression model to examine how governance conditions shape CSR effectiveness, and a difference-in-differences design to capture the impact of major CSR events. Data are drawn from brand finance; Interbrand; CSMAR; Alibaba's environmental, social, and governance (ESG) reports; and CSI-ESG ratings. Results indicate that about 45% of CSR's total effect on brand value is mediated through brand strength. Threshold patterns also appear: smaller boards amplify CSR's influence, and concentrated ownership strengthens its alignment with long-term reputation.

Open Access

View Full Issue

Exploring the Role of Digital Finance on Green Finance in the Chinese Context View Full Issue

This study examines the S&P Global Renewable Energy Index, the MSCI Global Markets Index, and the S&P Green Bond Index of China to investigate the complex relationship between green finance and... Show More

This study examines the S&P Global Renewable Energy Index, the MSCI Global Markets Index, and the S&P Green Bond Index of China to investigate the complex relationship between green finance and digital finance, and to demonstrate how these two mechanisms impact the adoption of renewable energy. This study further examines how digital financial technologies, such as blockchain, mobile payment systems, and big data analytics, can facilitate the acquisition of green funds for renewable energy projects. The investigation shows that digital finance dramatically lowers renewable energy project funding costs. Digital tools have sped up procedures, decreased intermediaries, and boosted risk evaluation, saving money. More people could invest in green energy with cheaper money. The authors observed that digital finance affects sustainable funding differently by region. Digital currency development is slow in rural areas but fast in cities. Digital financing has considerably decreased renewable energy risk and increased transparency.

Open Access

View Full Issue

Optimization of Financial Risk Assessment Decision Framework: A Case Study Based on Credit Default Data of Small and Micro Enterprises View Full Issue

Small and micro enterprises (SMEs) play a critical role in economic development, yet their access to credit is often constrained by inadequate risk assessment frameworks. Traditional credit scoring... Show More

Small and micro enterprises (SMEs) play a critical role in economic development, yet their access to credit is often constrained by inadequate risk assessment frameworks. Traditional credit scoring models struggle to capture the non-linearity, feature sparsity, and class imbalance inherent in SME financial data. To address these challenges, the authors propose HECRO (Heterogeneous Ensemble Credit Risk Optimizer), a multi-layered framework that integrates kernel-based and heuristic feature selection, ensemble base learners, and a Bayesian-optimized meta-stacking classifier. HECRO leverages KFS-MCLOC and BOWOA-KS for robust feature extraction, followed by GA-BPNN, SMOTE-XGBoost, Wide & Deep, and BD-LR models as diverse predictors, culminating in a BO-XGBoost meta-learner. SHAP-based interpretation enhances post-hoc transparency. These results demonstrate HECRO's superiority in both predictive accuracy and robustness. The study offers a practical and scalable solution for SME credit evaluation, providing new insights into the design of intelligent financial risk assessment systems.

Open Access

View Full Issue

Resolving Interpretability Conflicts of Gradient Boosted Decision Trees (XGBoost) in Credit Risk Assessment: Evidence From Internet Enterprise Finance Innovation A/B Testing View Full Issue

Credit risk assessment is a cornerstone of modern internet finance, yet the widespread adoption of machine learning models such as XGBoost is hindered by interpretability conflicts. Conventional... Show More

Credit risk assessment is a cornerstone of modern internet finance, yet the widespread adoption of machine learning models such as XGBoost is hindered by interpretability conflicts. Conventional boosting approaches often deliver strong predictive performance but produce inconsistent explanations across global and local levels, and they frequently violate domain-specific constraints such as monotonicity and fairness. To address these challenges, the authors propose UniXGB, a Unified Conflict-Resolving Explainable Gradient Boosting framework that integrates a generalized extreme value (GEV) boosting backbone with a consistency layer, multi-view explainability fusion, a conflict correction module, and an actionable counterfactual layer. The framework further supports federated adaptation and is validated through large-scale internet finance A/B testing. These findings confirm that UniXGB bridges the gap between accuracy and interpretability, offering a practical and trustworthy solution for deploying machine learning models in regulated financial environments.

Open Access

View Full Issue

A Case Study on Mitigating Financial Biases in M&A Target Through Intelligent Algorithmic Approaches View Full Issue

Mergers and acquisitions (M&A) are a cornerstone of corporate strategy, yet target valuation is often skewed by systematic biases that degrade judgment, misallocate capital, and reduce deal quality.... Show More

Mergers and acquisitions (M&A) are a cornerstone of corporate strategy, yet target valuation is often skewed by systematic biases that degrade judgment, misallocate capital, and reduce deal quality. Machine-learning models boost point accuracy but falter under temporal drift, spurious dependencies and sector anomalies, failing to meet needs of organizational investors and financial practitioners in dynamic markets. This study introduces the Bias-Corrected Intelligent Valuation Framework (BCIVF), an integrated system with bias-detection layer, adaptive correction mechanism, temporal dynamics encoder, and industry-network encoder. Across 5 complementary datasets, it outperforms 7 baselines: cuts MAE by 14.6%, raises bias-reduction ratio to 0.61, improves risk-adjusted decision scores from 0.44 to 0.52. Robust across tech, healthcare, manufacturing—key domains for organizational M&A activity; case studies show its valuations align with real deals, enhancing decision utility for organizations and advancing intelligent financial computing for end users.

Open Access

View Full Issue

A Gannet-Optimized Discrete Hidden Markov Framework for Investment Support Systems View Full Issue

Organizations operate under fixed budgets, yet latent market regimes dictate shifting capital and risk limits. Dynamically labeling these regimes from public prices is therefore critical. However... Show More

Organizations operate under fixed budgets, yet latent market regimes dictate shifting capital and risk limits. Dynamically labeling these regimes from public prices is therefore critical. However, the task is discrete and poorly served by continuous models. The authors employ discrete hidden Markov models (DHMMs) to infer categorical states from continuous price series. On the other hand, real-world data are non-stationary and noisy, and the traditional Baum-Welch routine often stalls in local optima, inflating runtime and causing overfitting. To overcome these limitations, they introduce GOA-DHMM, the first adaptation of the Gannet Optimization Algorithm to DHMMs. Hybrid encoding plus temporal regularization enforces probability-simplex constraints while preserving sequential coherence. Evaluated on 1,100 days of CSI-300 data, GOA-DHMM outperforms BW, PSO, and GA in accuracy and stability, delivering clearer, faster regime signals that let organizations allocate capital and hedge risk with confidence.

Open Access

View Full Issue

From Digital Intent to Resilient Outcomes: The Sequential Role of Smart Technologies and Green Supply Chain Management View Full Issue

The study empirically validates firm resilience (FR) as the outcome of a sequential capability-building process. The authors challenged the conventional literature by arguing that FR is a direct... Show More

The study empirically validates firm resilience (FR) as the outcome of a sequential capability-building process. The authors challenged the conventional literature by arguing that FR is a direct by-product of Digital Transformation, Smart Technologies (ST), and Green Supply Chain Management (GSCM), and integrates them into a sequential pathway. They used 306 logistics firms in China for analysis. They employed the Partial Least Squares Structural Equation Modeling (PLS-SEM) technique. The study finds that DT's impact on FR is only through sequential mediation. Moreover, DT leads to ST adoption, which significantly impacts GSCM practices that ultimately lead to enhanced FR. In addition, IP significantly moderates the ST–GSCM relationship. The study provides new insights into resilience and in the emerging context of China by integrating DCV and institutional theory. The study explains the impact of DT on FR through a sequential mediation approach that contributes to both theoretical and empirical research.

Open Access

View Full Issue

Application Analysis of Multi-Agent Collaborative AI-Driven Models in Urban Emergency Management: Model Construction for Urban Emergency Management View Full Issue

With global urbanization on the rise, cities face complex cascading risk impacts. Traditional emergency management has bottlenecks like fragmented multi-agent collaboration and single-point AI... Show More

With global urbanization on the rise, cities face complex cascading risk impacts. Traditional emergency management has bottlenecks like fragmented multi-agent collaboration and single-point AI application, making it hard to respond to dynamic emergencies. This study integrates Complex Adaptive System (CAS) theory, AI empowerment theory, and AnyLogic simulation validation to develop a multi-agent collaborative AI-driven model. It builds an “agent-data-technology-process” four-layer coupled architecture, clarifies multi-agent interaction rules and AI integration paths, and covers the full “prevention-preparedness-response-recovery” cycle. Empirical validation shows that in three typical scenarios—urban waterlogging, large-scale crowd gatherings, cross-regional public health incidents—vs. traditional models, the proposed model shortens response time, improves resource efficiency, and reduces casualties. It resolves emergency collaboration dilemmas, enriches AI-empowered emergency management's theoretical system, and guides building a resilient smart city emergency system.

Open Access

View Full Issue

The Effects of AI Attributes and Parasocial Relationships on Behavioral Intention Toward Virtual Influencers View Full Issue

This study examines the extent to which AI applicants such as virtual influencers develop parasocial relationships (PSRs) with people and how this relationship influences utilitarian and hedonic... Show More

This study examines the extent to which AI applicants such as virtual influencers develop parasocial relationships (PSRs) with people and how this relationship influences utilitarian and hedonic values, ultimately shaping behavioral intention. Furthermore, it investigates the direct and indirect effects of intelligence, and both external and internal anthropomorphism, on behavioral intention toward virtual influencers. To test the research model, survey data were analyzed using Partial Least Squares. The findings indicate that PSR directly influences behavioral intention and also affects it indirectly via utilitarian and hedonic values. External anthropomorphism enhances hedonic value, while internal anthropomorphism strengthens PSR. Moreover, intelligence significantly impacts PSR and both value types. This study contributes theoretically by clarifying how user perceptions relate to behavioral intentions toward AI applications and offers practical guidance for developers and marketers in designing effective interaction strategies.

Open Access

View Full Issue

A Study on Structural Transformation as Drivers of Industrial Upgrading: Emerging Markets Data Analysis Case View Full Issue

This study explores the transformative role of global value chain (GVC) participation in driving industrial upgrading in China, Vietnam, and India between 2018 and 2023. Employing an explanatory... Show More

This study explores the transformative role of global value chain (GVC) participation in driving industrial upgrading in China, Vietnam, and India between 2018 and 2023. Employing an explanatory, mixed-methods research design, it examines the causal relationships between GVC integration and structural economic transformations. This research thoroughly examines the smile curve, leapfrogging, and global production network theories, showing where they fit and their constraints regarding GVC relations in developing economies. The results provide insights to such stakeholders as policymakers, indicating that strategies aimed at GVC structural growth and competitiveness maximization should be implemented while advising against low-risk strategies such as being overly dependent, regionalized, or technology dependent. Offering the perspective of GVC participation, this study sheds light on the ongoing transformation of the world economy structure and industrial development in global competition.

Open Access

View Full Issue

Carbon Credit Market Pricing in a Big Data Financial Environment: A Bayesian Estimation and Deep Reinforcement Learning Approach View Full Issue

Global carbon markets face challenges in forecasting prices and aligning policies due to volatility, regulatory constraints, and uncertainty. While existing models improve predictions, they fail to... Show More

Global carbon markets face challenges in forecasting prices and aligning policies due to volatility, regulatory constraints, and uncertainty. While existing models improve predictions, they fail to incorporate uncertainty into decision-making. To address this, the authors propose a Bayesian Deep Reinforcement Learning framework for Carbon Pricing, which combines probabilistic price modeling, uncertainty propagation, and constraint-based reinforcement learning. Experiments across five datasets show that the model reduces error by 28%, improves performance by 25%, and boosts trading profit by 22%, while maintaining high emissions compliance. Stress-test results confirm the model's robustness, demonstrating that uncertainty-aware learning enhances the stability and efficiency of carbon credit markets.

Open Access

View Full Issue

DL-FinRisk: Deep Learning Model for Financial Enterprise Risk Assessment Under Data Legal Framework View Full Issue

To address existing financial risk assessment models' poor adaptability to complex scenarios, weak multi-source data integration, and inadequate dynamic threat response, this study proposes... Show More

To address existing financial risk assessment models' poor adaptability to complex scenarios, weak multi-source data integration, and inadequate dynamic threat response, this study proposes DL-FinRisk, a deep learning-driven framework incorporating Data Law-oriented compliance considerations. It integrates multi-modal fusion for heterogeneous data unification, a ResNet-LSTM hybrid architecture for spatial-temporal feature extraction, Bayesian network-based Dynamic Security Assessment (DSA) for real-time risk updates, and TOPSIS for decision-making. Validated on real-world financial data and public datasets under Data Law and regulatory compliance constraints, DL-FinRisk achieves 95.7% accuracy, 94.0% F1-score, and 73.1% Risk Reduction Rate, outperforming baseline models.

Open Access

View Full Issue

Dynamic Pricing Strategy Optimization Based on a Reinforcement Learning PPO Algorithm: An Empirical Study on Ride-Hailing Platforms View Full Issue

Dynamic pricing has become a cornerstone of ride-hailing platforms, yet designing strategies that simultaneously maximize revenue, ensure fairness, and maintain operational efficiency remains a... Show More

Dynamic pricing has become a cornerstone of ride-hailing platforms, yet designing strategies that simultaneously maximize revenue, ensure fairness, and maintain operational efficiency remains a formidable challenge. Traditional reinforcement learning approaches often optimize a single dimension—such as profitability or fairness—at the expense of others, limiting their applicability in real-world markets. To address this gap, the authors propose PRIME-PPO (Pricing with Repositioning Integration, Mechanism-awareness, and Equity via Proximal Policy Optimization), a unified reinforcement learning framework tailored for dynamic pricing. PRIME-PPO extends the PPO backbone with five innovations: a primal–dual mechanism for fairness and budget enforcement, mechanism-aware action masking to preserve incentive compatibility, auxiliary signals from dispatch and repositioning tasks to capture system-level dynamics, hierarchical grouping with parameter sharing for scalability, and dual-critic value distillation from TD3 and DQN for improved sample efficiency.

Open Access

View Full Issue

AI-Driven Service Industry Employee Behavior Prediction and Performance Optimization: Evidence From the Banking Industry View Full Issue

Behavioral, operational, and contextual factors affecting employee performance in the service industry are complex, and hence, predicting and intervening on them are especially difficult. As the... Show More

Behavioral, operational, and contextual factors affecting employee performance in the service industry are complex, and hence, predicting and intervening on them are especially difficult. As the banking industry is a crucial component of the service sector, it is particularly important to focus on predicting and intervening in the factors affecting the performance of banking employees. Current AI-based HR analytics solutions do not have a system for preserving fairness and optimizing decisions, as they tend to be based on a single source of signals. To solve these challenges, the present study introduces BEBOP-Net, a composite multi-source behavioral modeling and reinforcement learning agent that is used to elucidate expected employee behaviors and provide managerial control interventions that are comparatively inexpensive to implement. The structured HR features, chronological logs of the operational history, and written interaction records are modeled using a single multi-source behavior encoder with high temporal attention to extract dynamic patterns.

Open Access

View Full Issue

A Research Framework for Controllable Multimodal Visual Design Generation View Full Issue

Controllable image generation has advanced rapidly with diffusion models, yet existing approaches struggle to integrate multiple heterogeneous control signals and lack mechanisms for enforcing... Show More

Controllable image generation has advanced rapidly with diffusion models, yet existing approaches struggle to integrate multiple heterogeneous control signals and lack mechanisms for enforcing professional design principles. Addressing this challenge requires a unified framework capable of harmonizing semantic, structural, and stylistic modalities while maintaining geometric precision and aesthetic coherence. This study proposes UMC-Design, a unified multimodal controllable framework that introduces a shared control representation, a cross-domain fusion mechanism, and a dual-path diffusion architecture synchronized through a design prior network. The model jointly processes text, vector layouts, semantic maps, and reference images, enabling flexible and scalable multimodal conditioning. Experiments on COCO-Stuff, RICO + PubLayNet, and Crello demonstrate that UMC-Design achieves state-of-the-art performance, reducing FID to 22.3 and improving multimodal alignment to 0.81, surpassing leading baselines by large margins.

Open Access

View Full Issue

AI-Driven Public Health Information Management and Emergency Decision-Making: A Case Study of Hospital Information Systems View Full Issue

The increasing complexity of healthcare systems and the rapid growth of heterogeneous medical data pose significant challenges to effective decision-making in public health and clinical practice.... Show More

The increasing complexity of healthcare systems and the rapid growth of heterogeneous medical data pose significant challenges to effective decision-making in public health and clinical practice. Existing data-driven approaches often struggle to balance predictive accuracy, robustness, and interpretability, particularly under dynamic and uncertain conditions. To address these challenges, this study proposes an AI-driven Emergency Decision-Making framework (AIM-EDM) that integrates multi-source health data, temporal modeling, and causal reasoning into a unified decision-support architecture. The proposed framework leverages deep representation learning to capture complex temporal patterns, incorporates knowledge-guided causal inference to enhance interpretability, and employs decision optimization to support reliable and actionable outcomes.

Open Access

View Full Issue

Data Asset Content Decision Coordination in Supply Chain Management: A Multi-Level Analysis Based on Cross-Industry Cases View Full Issue

Supply chains are increasingly multi-layered and interdependent, and this complexity makes coordination across echelons a persistent challenge. Local optimization often backfires: a retailer's... Show More

Supply chains are increasingly multi-layered and interdependent, and this complexity makes coordination across echelons a persistent challenge. Local optimization often backfires: a retailer's attempt to reduce stockouts can trigger larger fluctuations upstream, producing the well-known bullwhip effect. Earlier studies have provided partial responses—reinforcement learning improves local policies, contract theory offers theoretical incentive alignment, and system dynamics clarifies structural causes—but taken separately, these approaches fall short of resolving multi-level coordination under uncertainty. In this study, the authors introduce the Hybrid Multi-Level Coordination Framework (HMLCF), which brings together multi-agent reinforcement learning for adaptive decision-making, contractual mechanisms to align decentralized incentives, system-dynamics modules to capture lead-time pipelines, and interpretable models that distill policies into transparent decision rules.

Open Access

View Full Issue

Enhancing Corporate Green Innovation Under the Green Factory Policy: A Graph Neural Network Approach for Management Decision-Making View Full Issue

As sustainable industrial transformation accelerates, understanding how green policies shape enterprise innovation becomes essential, yet conventional econometric methods cannot capture how policy... Show More

As sustainable industrial transformation accelerates, understanding how green policies shape enterprise innovation becomes essential, yet conventional econometric methods cannot capture how policy effects spread across industries, regions, and supply chains. This study introduces the Graph-Based Framework for Green Innovation Network, which models multi-dimensional enterprise relations and applies relation-specific attention to reveal dominant diffusion channels. Using manufacturing data from China from 2019 to 2023, the framework surpasses three baseline methods in determination coefficient, innovation classification, and Policy Diffusion Index. Ablation studies confirm the roles of contrastive alignment and reconstruction, and results show that policy incentives and peer learning drive green innovation diffusion while offering interpretable indicators for policy and enterprise decisions.

Open Access

View Full Issue

Beyond Automation: Investigating the Influence of Information Privacy and Overload on Nepotism and Bribery in Kuwait's E-Governance View Full Issue

This study addresses the inconsistent understanding of how Information and Communication Technology (ICT) impacts deeply entrenched corruption, specifically nepotism and bribery, which impose... Show More

This study addresses the inconsistent understanding of how Information and Communication Technology (ICT) impacts deeply entrenched corruption, specifically nepotism and bribery, which impose significant global economic costs. Academic research is sparse on how e-government solutions mitigate these issues, particularly in regions like the Middle East where localized corruption, such as Kuwait's 'wasta' (nepotism), is pervasive. The authors investigate this using Kuwait's “Sahel” application, a unified e-government platform. Through a custom-designed user perception questionnaire, the research explores whether the availability and use of official IT applications help overcome these distinct forms of corruption. The authors hypothesize that poor informational privacy and information overload positively correlate with both nepotism and bribery. Furthermore, they posit that bribery can act as a catalyst for nepotism.

Open Access

View Full Issue

Achieving Sustainable Manufacturing Through Digital Leadership: The Mediating Roles of Knowledge Management and Resilience View Full Issue

This study investigates the organizational drivers of sustainability performance in the manufacturing sector and proposes and tests an integrated model grounded in dynamic capability theory. Using... Show More

This study investigates the organizational drivers of sustainability performance in the manufacturing sector and proposes and tests an integrated model grounded in dynamic capability theory. Using structural equation modeling on empirical data, it examines the role of digital leadership as a catalyst for sustainable practices, mediated by digital knowledge management and organizational resilience. The findings confirm that digital leadership exerts a significant total effect on sustainability outcomes. Crucially, this influence is primarily channeled through two parallel mediating pathways: by building organizational resilience and by establishing systematic digital knowledge management. The model demonstrates high explanatory power, revealing that sustainability performance is a systemic achievement, dependent on the synergistic interaction of strategic leadership, institutionalized learning, and adaptive capacity.

Open Access

View Full Issue

Application of Natural Language Processing-Based User Emotion and Behavior Recognition in Enterprise Service Design Management View Full Issue

In enterprise service management, accurately recognizing user emotions and behaviors from natural language interactions is essential for supporting effective service routing, prioritization, and... Show More

In enterprise service management, accurately recognizing user emotions and behaviors from natural language interactions is essential for supporting effective service routing, prioritization, and escalation. However, existing approaches often treat emotion recognition and behavior prediction as independent tasks, overlooking their dynamic interaction across multi-turn service dialogues, which limits their effectiveness in emotionally intensive and long-horizon scenarios. To address this challenge, the authors propose SERVE-Net, a unified framework that explicitly models emotion–behavior interactions for enterprise service management. SERVE-Net jointly learns emotional dynamics, behavioral intent transitions, and service action decisions within a single architecture. Overall, this work shows that explicitly modeling emotion–behavior interactions is critical for bridging affective computing and enterprise service management, enabling intelligent service systems that are both accurate and operationally meaningful.

Open Access

View Full Issue

Large Language Model–Enabled Decision Support for Complex Project Management: A Case Study of Intelligent Urban Traffic Planning View Full Issue

The rapid development of intelligent urban traffic systems has transformed traffic planning into a complex project management problem characterized by multi-stage decision-making, uncertainty, and... Show More

The rapid development of intelligent urban traffic systems has transformed traffic planning into a complex project management problem characterized by multi-stage decision-making, uncertainty, and strict policy constraints. Although large language models (LLMs) have recently demonstrated strong reasoning and planning capabilities, existing approaches often lack robustness, regulatory compliance, and long-term decision stability when applied to real-world urban traffic planning scenarios. To address these challenges, this paper proposes PM-LLM, a lifecycle-aware and tool-augmented LLM framework designed to support complex project management decision-making in intelligent urban traffic planning. PM-LLM integrates constraint-aware knowledge retrieval, uncertainty scenario modeling, hierarchical decision decomposition, simulation-grounded risk-sensitive selection, and closed-loop revision into a unified Plan–Simulate–Operate–Revise decision loop.

Open Access

View Full Issue

Digital Leadership and Sustainability Outcomes: A Moderated Mediation Perspective From Manufacturing Organizations View Full Issue

While the manufacturing industry may use information technology to reduce adverse environmental impact, it needs related managerial staff who possess specialized knowledge to optimize sustainability... Show More

While the manufacturing industry may use information technology to reduce adverse environmental impact, it needs related managerial staff who possess specialized knowledge to optimize sustainability performance and high-resilience organizations that have flexible hierarchies for changes. Yet, few studies have studied this knowledge and the impact of organizational resilience on firms' sustainability in a holistic way. To fill this academic research gap, this study surveyed 307 managers, IT executives, and sustainability manufacturing industry officers from five major industrial provinces in China. The partial least squares structural equation model results indicated that digital leadership's direct impact on the sustainability performance of manufacturing firms was mediated through digital knowledge management. At the same time, low-resilience organizations rely more on digital knowledge management to enhance sustainability outcomes, but highly resilient organizations may already possess systems that support sustainability.

Open Access

View Full Issue

Latency Trade-Offs of a Lightweight YOLOv11 Detection Algorithm on Edge Computing Nodes for Factory Safety Decision-Making: An On-Site Experiment in Smart Manufacturing View Full Issue

Ensuring real-time safety in smart manufacturing environments increasingly relies on vision-based object detection systems deployed at the edge. However, achieving a balance between inference... Show More

Ensuring real-time safety in smart manufacturing environments increasingly relies on vision-based object detection systems deployed at the edge. However, achieving a balance between inference accuracy and low-latency performance remains a key challenge, particularly when deploying deep learning models on resource-constrained industrial edge devices. To address this, the authors propose Factory-Aware YOLOv11(FA-YOLOv11), a lightweight object detection framework designed for latency-aware safety decision-making in factory settings. The method introduces a pruned and quantized YOLOv11 backbone, augmented with multi-scale feature fusion and a latency-gated decision mechanism to suppress unstable detections. Temporal filtering further enhances precision under noisy conditions. This study highlights the importance of latency–precision co-optimization for edge intelligence in smart factories and provides a deployable solution to improve factory safety through fast and accurate perception.

Open Access

View Full Issue

A Deep Learning-Based System for Automatic Classification and Management of Museum Artifacts View Full Issue

The rapid digitization of museum collections demands intelligent systems capable of automatically classifying and managing large-scale artifact inventories. Existing methods often address... Show More

The rapid digitization of museum collections demands intelligent systems capable of automatically classifying and managing large-scale artifact inventories. Existing methods often address classification and retrieval separately and fail to satisfy practical management requirements such as multimodal integration and system-level consistency. A deep learning-based system is proposed for the automatic museum classification and management, which integrates artifact localization, robust visual representation learning, multimodal semantic fusion, and management-oriented indexing within a single framework. Experiments conducted on large public museum datasets demonstrate that the proposed system outperforms existing methods in classification accuracy, reliability, and end-to-end automation performance. These results highlight the effectiveness of a unified, management-aware design in real-world museum applications.

Open Access

View Full Issue

LLM-Driven Business Intelligence for Retail Digital Transformation: A Decision Support System Case Study View Full Issue

The rapid digital transformation of the retail industry has generated increasing demand for intelligent business intelligence systems that can translate natural language managerial inquiries into... Show More

The rapid digital transformation of the retail industry has generated increasing demand for intelligent business intelligence systems that can translate natural language managerial inquiries into actionable analytical and predictive decisions. However, existing enterprise analytics solutions either rely on rigid dashboard-based querying or isolated AI models, and thus struggle to jointly achieve knowledge grounding, executable analytical reasoning, and decision-oriented synthesis within a unified framework. RIDE constructs knowledge-grounded analytical contexts from enterprise repositories, employs multi-step reasoning–action planning to orchestrate tool-based analytics, generates grammar-constrained executable SQL for reliable database access, and integrates predictive forecasting to support decision synthesis. Overall, this study establishes a new paradigm for large language model–driven business intelligence systems that bridge human-like inquiry and machine-executable analytics, providing a robust foundation for intelligent and trustworthy retail digital management.

Open Access

View Full Issue

An Optimization-Driven Approach for Accurate Prediction of Diabetes Mellitus View Full Issue

Diabetes mellitus is a progressive metabolic disorder requiring timely identification to prevent severe complications and reduce healthcare burdens. This paper proposes an optimization-driven deep... Show More

Diabetes mellitus is a progressive metabolic disorder requiring timely identification to prevent severe complications and reduce healthcare burdens. This paper proposes an optimization-driven deep learning framework for accurate diabetes prediction, integrating Long Short-Term Memory (LSTM) networks with Ant Colony Optimization (ACO) and Grey Wolf Optimizer (GWO) for feature selection and hyperparameter tuning. The framework is conceptualized as a decision-support system embedded within electronic health records and clinical workflows, supporting physicians, nurses, case managers, and patients. Key considerations—including technology acceptance, trust, transparency, workflow integration, governance, and privacy—are addressed to ensure alignment with real-world organizational contexts. The results demonstrate both technical feasibility and practical relevance, linking predictive analytics capability with human–AI collaboration, adoption factors, and operational decision-making in healthcare organizations.

Open Access

View Full Issue

Optimizing an Automated Storage and Retrieval System Using a Firefly and Simulation-Based Approach View Full Issue

Automated storage and retrieval systems (AS/RSs) play a crucial role in an enterprise. However, conventional AS/RSs often lack adaptability to complex production environments, particularly in... Show More

Automated storage and retrieval systems (AS/RSs) play a crucial role in an enterprise. However, conventional AS/RSs often lack adaptability to complex production environments, particularly in multi-floor manufacturing firms (MFMFs), where inter-floor movement of stock-keeping units (SKUs) poses significant challenges. Thus, this study proposes a novel hybrid metaheuristic, Hybrid (IFA+NN), for solving the storage/retrieval (S/R) machine scheduling problem (SRSP) in an aisle-captive AS/RS within an MFMF. The S/R machine uses dual commands to complete one storage and one retrieval task in a single round trip. The Hybrid (IFA+NN) combines an improved Firefly Algorithm (IFA) with a Nearest Neighbor (NN) heuristic to enhance both global search capability and local decision efficiency. Computational experiments show that Hybrid (IFA+NN) consistently outperforms benchmark methods, including FA, PSO, GA, and FCFS. The study examines SKU storage allocation strategies and demonstrates that jointly considering storage and usage positions further improves overall system performance.

Open Access

View Full Issue

A Deep Learning-Based Model for User Emotion Recognition and Behavior Guidance Design: EMoGuideNet - An Emotion-Aware Multimodal Framework for Enhancing Intelligent User Interaction View Full Issue

EMoGuideNet is a deep-learning model that integrates user emotion and behavior in multi-turn interactions, overcoming the current limitation of treating them as separate processes. By simultaneously... Show More

EMoGuideNet is a deep-learning model that integrates user emotion and behavior in multi-turn interactions, overcoming the current limitation of treating them as separate processes. By simultaneously streaming emotional conditions and behavioral responses, it captures their interactive dynamics. The model processes multimodal inputs (textual/acoustic) through a recurrent emotion tracking mechanism, enabling temporal emotion modeling, and uses multi-task learning to predict behavior. Evaluated on four benchmarks (MELD, IEMOCAP, DailyDialog, ReDial), it outperforms CoCoLM, SER-Meta, and BERT4Rec, achieving a statistically significant 3.9% improvement in emotion-recognition accuracy and 4.6% in behavioral Hit@10 on MELD (p < 0.01). The model also demonstrates robustness to noise and generalizes well to unseen settings, highlighting the effectiveness of emotion-conditioned behavior modeling. These advancements enhance adaptive dialogue systems by enabling responses with greater emotional relevance and situational awareness.

Open Access

View Full Issue

Cyber Crime Risk Classification and Anomaly Detection Based on Self-Supervised Learning View Full Issue

The proliferation of digital services has led to a surge in cyber crimes, while traditional detection methods face bottlenecks such as labeled data scarcity, data imbalance, and poor adaptability to... Show More

The proliferation of digital services has led to a surge in cyber crimes, while traditional detection methods face bottlenecks such as labeled data scarcity, data imbalance, and poor adaptability to zero-day attacks. To address these issues, this study proposes CRAD-Defender, an end-to-end framework integrating self-supervised learning (SSL), graph neural networks (GNNs), and reinforcement learning (RL). It fuses MoCo v3 and GraphSAGE-GAT to capture individual behavioral features and cross-entity relational risks, adopts One-Class SVM for initial anomaly detection, and leverages DQN for closed-loop optimization of detection parameters. Experimental evaluations on UNSW-NB15 and CSE-CIC-IDS2019 datasets show CRAD-Defender outperforms baseline models with an F1-score of 0.91 and 0.89, ultra-low FPR of 1.1% and 1.3%, and low detection latency. Ablation studies verify the synergistic effectiveness of each core module, demonstrating the framework's superiority in accurate, real-time cyber crime risk classification and anomaly detection.

Open Access

View Full Issue

A Generalized Deep Learning Framework for Robust Medical Image Segmentation View Full Issue

Semantic segmentation is crucial for medical image analysis, enabling accurate delineation of anatomical structures and pathological regions. However, heterogeneity across imaging modalities such as... Show More

Semantic segmentation is crucial for medical image analysis, enabling accurate delineation of anatomical structures and pathological regions. However, heterogeneity across imaging modalities such as ultrasound, fundus photography, X-ray, and computed tomography (CT) challenges model generalization. This study proposes GenMed-Net, a generalized segmentation framework based on a U-Net backbone integrating a ResNet50 encoder, a Spatial Channel Block Attention Module (SCBAM), and an Atrous Spatial Pyramid Pooling (ASPP) module. GenMed-Net is evaluated on four heterogeneous datasets: thyroid ultrasound, retinal fundus vessel images, pulmonary fibrosis chest X-rays, and the CC-CCII pneumonia chest CT dataset. The model achieves Dice Similarity Coefficients of 91.87%, 96.15%, 98.99%, and 89.11%, respectively, outperforming representative CNN- and Transformer-based methods. Visual and attention heatmap analyses further demonstrate improved lesion localization and strong cross-modality generalization, supporting its potential for clinical decision support.

Open Access

View Full Issue

Institutional Pressure, Digital Leadership, and ESG Integration: A Moderated Mediation Model of Organizational Sustainability View Full Issue

This study investigates the strategic pathways through which manufacturing firms achieve organizational sustainability under intense institutional pressure. Using a positivist, quantitative... Show More

This study investigates the strategic pathways through which manufacturing firms achieve organizational sustainability under intense institutional pressure. Using a positivist, quantitative approach, the authors collected survey data from 355 experienced managers and professionals across key manufacturing provinces and industries in China. Analysis via structural equation modeling provides robust support for the hypothesized model. Results confirm that ESG Integration Capability fully mediates the relationship between Digital Leadership and Organizational Sustainability. Crucially, Institutional Pressure significantly moderates the first stage of this mediation: the effect of Digital Leadership on ESG Integration Capability is substantially stronger under higher perceived pressure. This finding reveals that in China's policy-driven ecosystem, institutional forces act not as a direct driver but as a contextual amplifier that potentiates leadership effectiveness. The model explains 40.4% of the variance in sustainability performance, demonstrating strong predictive relevance.