Search Results (947)

Driven by the rapid evolution of AI technology, compatible management mechanisms have become a systematic project involving the participation of multiple stakeholders. However, constrained by the rigidity and lag of traditional laws, the “one-size-fits-all” regulatory model will exacerbate the vulnerability of the complex system of AI governance, hinder the sustainable evolution of the AI ecosystem that relies on the dynamic balance between innovation and responsibility, and ultimately fall into the dilemma of “chaos when laissez-faire, stagnation when over-regulated”. To address this challenge, this study takes the multi-stakeholder collaborative mechanism co-established by governments, enterprises, and third-party technical audit institutions as its research object and centers on the issue of “strategic fluctuations” caused by key factor disturbances. From the perspective of the full life cycle of technological development, the study integrates the historical compliance performance of stakeholders and develops a nonlinear dynamic reward and punishment mechanism based on Credit Entropy. Through evolutionary game simulation, it further examines this mechanism as a realization path to promote the transformation from passive campaign-style AI supervision to agile governance of AI, which is characterized by rapid response and minimal intervention, thereby laying a foundation for the sustainable development of AI technology that aligns with long-term social well-being, resource efficiency, and inclusive growth. Finally, the study puts forward specific governance suggestions, such as setting access thresholds for third-party institutions and strengthening their independence and professionalism, to ensure that the iterative development of AI makes positive contributions to the sustainability of socio-technical systems. Full article

For more than a decade (2014–2025), the Consortium “3D for Digital Humanities” has been advancing the use of 3D technologies in the Humanities and Social Sciences (HSS) while structuring and supporting the research community. It now brings together more than 30 teams, primarily from academic research, but also increasingly from the cultural sector. Under its coordination, significant achievements have been realized, including best-practice guides, an infrastructure for the publication of 3D data, and dedicated software for documentation, dissemination, and archiving, as well as a metadata schema, all fully aligned with FAIR principles. The Consortium has developed national training programs, particularly on metadata and ethical practices, and contributed to important initiatives such as the reconstruction of Notre-Dame de Paris, while actively engaging in European projects. It has also fostered international collaborations to broaden perspectives, share methodologies, and amplify impacts. Looking ahead (2025–2033), the Consortium aims to address the environmental impact of 3D data production and storage by proposing best practices for digital sustainability and efficiency. It is also expanding the National 3D Data Repository, enhancing interoperability, and adopting emerging standards to meet evolving scientific needs. Building on its past achievements, the Consortium intends to further advance 3D research and its applications across disciplines, positioning 3D data as a key component of future scientific data clouds. Full article

Against the backdrop of the world’s increasing reliance on renewable energy, the inherent intermittency and volatility of wind and solar energy pose significant challenges to the stability and economic benefits of the power system. In regions rich in renewable energy resources such as Gansu Province, due to low operational efficiency and underdeveloped market mechanisms, the potential of new energy storage systems is often not fully exploited. This paper proposes an integrated shared energy storage model designed to suppress wind power fluctuations and a two-way market trading mechanism designed to maximize social welfare to solve these problems. Firstly, a hybrid energy storage system combining electrochemical- and hydrogen-based energy storage is constructed. The modular coordination strategy is adopted to dynamically allocate power capacity, and the wind energy fluctuation suppression technology is proposed to achieve fluctuation suppression at multiple time scales. Secondly, a combined dual bidding mechanism is introduced, allowing for combined bidding across time periods and resource types, to better capture user preferences and enhance market flexibility. The model is represented as a mixed-integer nonlinear programming problem aimed at maximizing social welfare, and then transformed into a linear equivalence problem to enhance the traceability of the calculation. The branch and bound algorithm is adopted to solve this problem. Finally, the simulation results based on the bidding data of a certain area enhanced the participation of participants and improved the fairness of the market and the overall social welfare. This system effectively enhances the grid-friendliness of renewable energy grid connection and provides a scalable and replicable framework for highly renewable energy systems. Full article

The advent of large-scale AI models (LAMs) marks a pivotal shift in technological innovation with profound societal implications. While demonstrating unprecedented potential to enhance human well-being by fostering efficiency and accessibility in critical domains like medicine, agriculture, and education, their rapid deployment presents a double-edged sword. This progress is accompanied by significant, often under-examined, sustainability costs, including large environmental footprints, the risk of exacerbating social inequities via algorithmic bias, and challenges to economic fairness. This paper provides a balanced and critical review of LAMs’ applications across five key livelihood domains, viewed through the lens of sustainability science. We systematically analyze the inherent trade-offs between their socio-economic benefits and their environmental and social costs. We conclude by arguing for a paradigm shift towards ‘Sustainable AI’ and provide actionable, multi-stakeholder recommendations for aligning artificial intelligence with the long-term goals of a more equitable, resilient, and environmentally responsible world. Full article

The rapid growth of digital communication has intensified spam-related threats, including phishing and malware, which employ advanced evasion tactics. Traditional filtering methods struggle to keep pace, driving the need for sophisticated machine learning (ML) solutions. The effectiveness of ML models hinges on selecting high-quality input features, especially in high-dimensional datasets where irrelevant or redundant attributes impair performance and computational efficiency. Guided by principles of symmetry to achieve an optimal balance between model accuracy, complexity, and interpretability, this study proposes an Enhanced Hybrid Quantum-Inspired Firefly and Artificial Bee Colony (EHQ-FABC) algorithm for feature selection in spam detection. EHQ-FABC leverages the Firefly Algorithm’s local exploitation and the Artificial Bee Colony’s global exploration, augmented with quantum-inspired principles to maintain search space diversity and a symmetrical balance between exploration and exploitation. It eliminates redundant attributes while preserving predictive power. For interpretability, Shapley Additive Explanations (SHAPs) are employed to ensure symmetry in explanation, meaning features with equal contributions are assigned equal importance, providing a fair and consistent interpretation of the model’s decisions. Evaluated on the ISCX-URL2016 dataset, EHQ-FABC reduces features by over 76%, retaining only 17 of 72 features, while matching or outperforming filter, wrapper, embedded, and metaheuristic methods. Tested across ML classifiers like CatBoost, XGBoost, Random Forest, Extra Trees, Decision Tree, K-Nearest Neighbors, Logistic Regression, and Multi-Layer Perceptron, EHQ-FABC achieves a peak accuracy of 99.97% with CatBoost and robust results across tree ensembles, neural, and linear models. SHAP analysis highlights features like domain_token_count and NumberOfDotsinURL as key for spam detection, offering actionable insights for practitioners. EHQ-FABC provides a reliable, transparent, and efficient symmetry-aware solution, advancing both accuracy and explainability in spam detection. Full article

Symmetry is an essential characteristic in both solution spaces and cloud task scheduling loads, as it reflects a structural balance that can be exploited to enhance algorithmic efficiency and robustness. In recent years, with the rapid development of 6G networks, the number of tasks requiring computation in the cloud has surged, prompting an increasing number of researchers to focus on how to efficiently schedule these tasks to idle computing nodes at low cost to enhance system resource utilization. However, developing reliable and cost-effective scheduling schemes for cloud computing tasks in real-world environments remains a significant challenge. This paper proposes a method for cloud computing task scheduling in real-world environments using an improved dhole optimization algorithm (IDOA). First, we enhance the quality of the initial population by employing a uniform distribution initialization method based on the Sobol sequence. Subsequently, we further improve the algorithm’s search capabilities using a sine elite population search method based on adaptive factors, enabling it to more effectively explore promising solution spaces. Additionally, we propose a random mirror perturbation boundary control method to better address individual boundary violations and enhance the algorithm’s robustness. By explicitly leveraging symmetry characteristics, the proposed algorithm maintains balanced exploration and exploitation, thereby improving convergence stability and scheduling fairness. To evaluate the effectiveness of the proposed algorithm, we compare it with nine other algorithms using the IEEE CEC2017 test set and assess the differences through statistical analysis. Experimental results demonstrate that the IDOA exhibits significant advantages. Finally, to verify its applicability in real-world scenarios, we applied IDOA to cloud computing task scheduling problems in actual environments, achieving excellent results and successfully completing cloud computing task scheduling planning. Full article

In many developing regions, recruitment still relies heavily on traditional methods that often ignore the importance of aligning a candidate’s personality with the job role. This mismatch can lead to poor performance, dissatisfaction, and high turnover. To address this, the study presents a personality-aware recommendation system that combines the Myers–Briggs Type Indicator (MBTI) with machine learning to support smarter hiring decisions. The system is tailored for the South Asian job market and includes two main components: a web-based MBTI assessment for applicants and a dashboard for HR professionals powered by a XGBoost classifier. This model was trained on a dataset correlating applicant profiles and the flagged preferences of MBTI with the job. Experience and the number of skills, education level, and encoded MBTI types were the key features, and the SMOTE method was employed to balance the dataset. The model attained an accuracy of 74.30%, having balanced precision and recall measures. It was also discriminative, the ROC AUC was 0.84, and the precision–recall AUC was 0.85. One example of utilizing the Software Developer position in real life demonstrated the success of the system to filter and rank candidates at the same time according to both technical and personality-specific criteria. Overall, this study emphasizes the worth of combining insights from psychological profiling with machine learning in order to develop a more holistically, fair, and efficient hiring process. Full article

The advent of sixth-generation (6G) wireless networks and cell-free massive multiple-input multiple-output (MIMO) architectures underscores the need for efficient resource allocation to support federated learning (FL) at the network edge. Existing approaches often treat communication, computation, and learning in isolation, overlooking dynamic heterogeneity and fairness, which leads to degraded performance in large-scale deployments. To address this gap, we propose a joint optimization framework that integrates communication–computation co-design, fairness-aware aggregation, and a hybrid strategy combining convex relaxation with deep reinforcement learning. Extensive experiments on benchmark vision datasets and real-world wireless traces demonstrate that the framework achieves up to 23% higher accuracy, 18% lower latency, and 21% energy savings compared with state-of-the-art baselines. These findings advance joint optimization in federated learning (FL) and demonstrate scalability for 6G applications. Full article

This study investigates hotel employees’ perceptions of AI-driven human resource (HR) management systems within the Accor Group’s properties across three major European cities: Paris, Berlin, and Amsterdam. These diverse urban contexts, spanning a broad portfolio of hotel brands from luxury to economy, provide a rich setting for exploring how AI integration affects employee attitudes and work–life balance. A total of 437 employees participated in the survey, offering a robust dataset for structural equation modeling (SEM) analysis. Exploratory factor analysis identified two primary factors shaping perceptions: AI Perceptions, which encompasses employee views on AI’s impact on job performance, communication, recognition, and retention, and balanced management, reflecting attitudes toward fairness, personal consideration, productivity, and skill development in AI-managed environments. The results reveal a complex but optimistic view, where employees acknowledge AI’s potential to enhance operational efficiency and career optimism but also express concerns about flexibility loss and the need for human oversight. The findings underscore the importance of transparent communication, contextual sensitivity, and continuous training in implementing AI systems that support both organizational goals and employee well-being. This study contributes valuable insights to hospitality management by highlighting the relational and ethical dimensions of algorithmic HR systems across varied organizational and cultural settings. Full article

The rapid development of generative artificial intelligence (GenAI) has revolutionized how individuals and organizations interact with technology. These systems, ranging from conversational agents to creative tools, are increasingly embedded in daily life. However, their effectiveness relies heavily on the usability of their graphical user interfaces (GUIs), which serve as the primary medium for user interaction. Moreover, the design of these interfaces must align with ethical principles such as transparency, fairness, and user autonomy to ensure responsible usage. This study evaluates the usability of GUIs for three widely-used GenAI applications, including ChatGPT (GPT-4), Gemini (1.5), and Claude (3.5 Sonnet), using a heuristics-based and user-based testing approach (experimental-qualitative investigation). A total of 12 participants from a research organization in Australia, participated in structured usability evaluations, applying 14 usability heuristics to identify key issues and ethical concerns. The results indicate that Claude’s GUI is the most usable among the three, particularly due to its clean and minimalistic design. However, all applications demonstrated specific usability issues, such as insufficient error prevention, lack of shortcuts, and limited customization options, affecting the efficiency and effectiveness of user interactions. Despite these challenges, each application exhibited unique strengths, suggesting that while functional, significant enhancements are needed to fully support user satisfaction and ethical usage. The insights of this study can guide organizations in designing GenAI systems that are not only user-friendly but also ethically sound. Full article

Accurate brain tumor classification from MRI is often constrained by limited labeled data. We systematically compare conventional machine learning, deep learning, and few-shot learning (FSL) for four classes (glioma, meningioma, pituitary, no tumor) using a standardized pipeline. Models are trained on the Kaggle Brain Tumor MRI Dataset and evaluated across dataset regimes (100%→10%). We further test generalization on BraTS and quantify robustness to resolution changes, acquisition noise, and modality shift (T1→FLAIR). To support clinical trust, we add visual explanations (Grad-CAM/saliency) and report per-class results (confusion matrices). A fairness-aligned protocol (shared splits, optimizer, early stopping) and a complexity analysis (parameters/FLOPs) enable balanced comparison. With full data, Convolutional Neural Networks (CNNs)/Residual Networks (ResNets) perform strongly but degrade with 10% data; Model-Agnostic Meta-Learning (MAML) retains competitive performance (AUC-ROC ≥ 0.9595 at 10%). Under cross-dataset validation (BraTS), FSL—particularly MAML—shows smaller performance drops than CNN/ResNet. Variability tests reveal FSL’s relative robustness to down-resolution and noise, although modality shift remains challenging for all models. Interpretability maps confirm correct activations on tumor regions in true positives and explain systematic errors (e.g., “no tumor”→pituitary). Conclusion: FSL provides accurate, data-efficient, and comparatively robust tumor classification under distribution shift. The added per-class analysis, interpretability, and complexity metrics strengthen clinical relevance and transparency. Full article

Trees are vital to both environmental health and human well-being. They purify the air we breathe, support biodiversity by providing habitats for wildlife, prevent soil erosion to maintain fertile land, and supply wood for construction, fuel, and a multitude of essential products such as fruits, to name a few. Therefore, it is important to monitor and preserve them to protect the natural environment for future generations and ensure the sustainability of our planet. Remote sensing is the rapidly advancing and powerful tool that enables us to monitor and manage trees and forests efficiently and at large scale. Statistical methods, machine learning, and more recently deep learning are essential for analyzing the vast amounts of data collected, making data the fundamental component of these methodologies. The advancement of these methods goes hand in hand with the availability of sample data; therefore, a review study on available high-resolution aerial datasets of trees can help pave the way for further development of analytical methods in this field. This study aims to shed light on publicly available datasets by conducting a systematic search and filter and an in-depth analysis of them, including their alignment with the FAIR—findable, accessible, interoperable, and reusable—principles and the latest trends concerning applications for such datasets. Full article

With the expected increase in drone operations, inter-operator fairness issues and congestion problems are expected to arise due to the strategic authorization approach mandated in European regulation. As an alternative, the proposed authorization method is based on a deferred authorization decision with multiple-priority classes that are gate-kept by a series of scarce flight tokens. In it, operators can guide the aerial traffic deconfliction process by indicating the criticality of each operation (i.e., selected priority class) based on their business logic and the available flight tokens. Scarce token distribution is performed by a centralized service following a fairness- or congestion-management policy defined by authorities. Also, geographical and temporal incentives can be considered using a 4D-dependent temporal airspace cost to compute the required number of tokens per flight. Results based on several simulation scenarios demonstrate the validity of the approach and its capability in prioritizing different operators’ behaviors (fairness management) or avoiding flight hotspots (congestion management). Overall, it is concluded that the proposed method is an efficient, fair, simple and scalable novel authorization process that can be integrated into the U-space ecosystem. Full article

Driven by China’s “dual carbon” strategy, concerns about channel fairness and green investment have become key frontier issues in supply chain management. This study focuses on a two-tier supply chain under a low-carbon background and innovatively incorporates both fairness concerns and green investment perspectives. It systematically explores the impact mechanisms of fairness concern coefficients and green investment levels on channel pricing and profit distribution across four scenarios: information symmetry vs. asymmetry and the presence vs. absence of channel encroachment. The simulation results reveal the following: (1) Under information symmetry and without channel encroachment, an increase in the retailer’s fairness concern significantly enhances its bargaining power and profit margin, while the supplier actively adjusts the wholesale price to maintain cooperation stability. (2) Channel encroachment and changes in information structure intensify the nonlinearity and complexity of profit distribution. The marginal benefit of green investment for supply chain members shows a diminishing return, indicating the existence of an optimal investment range. (3) The green premium is predominantly captured by the supplier, while the retailer’s profit margin tends to be compressed, and order quantity exhibits rigidity in response to green investment. (4) The synergy between fairness concerns and green investment drives dynamic adjustments in channel strategies and the overall profit structure of the supply chain. This study not only reveals new equilibrium patterns under the interaction of multidimensional behavioral factors but also provides theoretical support for achieving both economic efficiency and sustainable development goals in supply chains. Based on these findings, it is recommended that managers optimize fairness incentives and green benefit-sharing mechanisms, improve information-sharing platforms, and promote collaborative upgrading of green supply chains to better integrate social responsibility with business performance. Full article

Connectivity is one of the essential needs in today’s standards in many aspects of life, starting with personal relationships, education, and remote work and ending with the security and economy of countries. However, connectivity is susceptible to intentional and unintentional disruptions, leading to great impact on critical infrastructures. Hence, maintaining connectivity is a crucial task to sustain the continuous flow of life. The challenge is to find an optimal recovery plan to reconnect all demands as soon as possible after the disruptive event, ensuring fairness in the process of reallocating the remaining resources. In this paper, we present a post-disruption recovery framework for networked systems to optimize the recovery plan to reconnect the network demands as soon as possible. More specifically, we introduce an algorithmic approach using a mathematical programming model that optimally recovers the disrupted arcs of the network while ensuring the highest connectivity. The proposed approach considers both fairness and efficiency through finding the MMF (max-min fairness) resource allocation throughout the recovery process. The proposed approach is tested on a variety of benchmark networks under a set of disruption levels; then, the results are compared with the maximum-flow model. Full article