Search Results (5,153)

With the globalized deployment of cross-border vehicle location services and the trajectory data, which contain user identity information and geographically sensitive features, the variability in privacy regulations in different jurisdictions can further exacerbate the technical and compliance challenges of data privacy protection. Traditional static differential privacy mechanisms struggle to accommodate spatiotemporal heterogeneity in dynamic scenarios because of the use of a fixed privacy budget parameter, leading to wasted privacy budgets or insufficient protection of sensitive regions. This study proposes a reinforcement-learning-based dynamic noise optimization method (DNO-RL) that dynamically adjusts the Laplacian noise scale by real-time sensing of vehicle density, region sensitivity, and the remaining privacy budget via a deep Q-network (DQN), with the aim of providing context-adaptive differential privacy protection for cross-border vehicle location services. Simulation experiments of cross-border scenarios based on the T-Drive dataset showed that DNO-RL reduced the average localization error by 28.3% and saved 17.9% of the privacy budget compared with the local differential privacy under the same privacy budget. This study provides a new paradigm for the dynamic privacy–utility balancing of cross-border vehicular networking services. Full article

Background/Objectives: Dietary patterns have changed over time, characterising a process of nutritional transition that reflects socioeconomic and demographic inequalities among different populations. This study assessed changes in dietary consumption patterns and the associated social determinants, comparing two time periods in a sample of individuals from a state in the Northeast Region of Brazil. Methods: Data from the 2008–2009 and 2017–2018 Household Budget Survey for the state of Paraíba were analysed, totalling 951 and 1456 individuals, respectively. Foods were categorised according to the NOVA classification and compared based on sociodemographic and economic variables. To determine the factors that most strongly explain the contribution of each NOVA food group to the diet, beta regression analysis was conducted. Results: Differences were observed between the two periods regarding the dietary contribution of the NOVA food groups, with a decrease in consumption of unprocessed foods and an increase in ultra-processed foods. Living in urban areas, being an adolescent, and having an income above the minimum wage were associated with reduced intake of unprocessed foods in both periods. Additionally, being an adolescent and having more than eight years of schooling were associated with higher consumption of ultra-processed foods. Conclusions: The population under study showed changes in food consumption, reflecting a transition process that is occurring unevenly across socioeconomic and demographic groups, thereby reinforcing social inequalities. These findings can guide priorities in food and nutrition policies, highlighting the need for intervention studies to evaluate the effectiveness of such actions. Full article

This article presents a spatial analysis of projects selected by city residents and implemented in five successive editions (2015–2019) of the participatory budgeting in Częstochowa, Poland. The study examines the relationship between the type of hard projects (small investments in public infrastructure and landscaping) and the pre-existing characteristics of the land use of each district. Kernel density estimation and Spearman correlation analysis were used. The highest spatial density occurred in projects related to the modernization of roads and sidewalks, recreation, and greenery, indicating a relatively high number of proposals within or near residential areas. Key correlations included the following: (1) greenery projects were more common in districts lacking green areas; (2) recreational infrastructure was more frequently chosen in areas with significant water features; (3) street furniture projects were mostly selected in districts with sparse development, scattered buildings, and postindustrial sites; (4) educational infrastructure was often chosen in low-density, but developing districts. The selected projects often reflect local deficits in specific land use or public infrastructure, but also stress the predestination of the recreational use of waterside areas. Full article

The strategic allocation of advertising billboards has become a critical aspect of urban planning and resource management. While previous studies have explored site selection based on road network and population data, they have often overlooked the diminishing marginal returns of overlapping coverage and neglected to efficiently process large-scale urban datasets. To address these challenges, this study proposes two complementary optimization methods: an enhanced greedy algorithm based on geometric modeling and spatial acceleration techniques, and a reinforcement learning approach using Proximal Policy Optimization (PPO). The enhanced greedy algorithm incorporates population-weighted road coverage modeling, employs a geometric series to capture diminishing returns from overlapping coverage, and integrates spatial indexing and parallel computing to significantly improve scalability and solution quality in large urban networks. Meanwhile, the PPO-based method models billboard site selection as a sequential decision-making process in a dynamic environment, where agents adaptively learn optimal deployment strategies through reward signals, balancing coverage gains and redundancy penalties and effectively handling complex multi-step optimization tasks. Experiments conducted on Wuhan’s road network demonstrate that both methods effectively optimize population-weighted billboard coverage under budget constraints while enhancing spatial distribution balance. Quantitatively, the enhanced greedy algorithm improves coverage effectiveness by 18.6% compared to the baseline, while the PPO-based method further improves it by 4.3% with enhanced spatial equity. The proposed framework provides a robust and scalable decision-support tool for urban advertising infrastructure planning and resource allocation. Full article

A shallow quaternary sedimentary aquifer within the river alluvial deposits of eastern Uzbekistan is increasingly vulnerable to the impacts of climate change and anthropogenic activities. Despite its essential role in supplying water for domestic, agricultural, and industrial purposes, the aquifer system remains poorly understood. This study employed a three-dimensional MODFLOW-based groundwater flow model to assess climate change impacts on water budget components under the SSP5-8.5 scenario for 2020–2099. Model calibration yielded RMSE values between 0.25 and 0.51 m, indicating satisfactory performance. Simulations revealed that lateral inflows from upstream and side-valley alluvial deposits contribute over 84% of total inflow, while direct recharge from precipitation (averaging 120 mm/year, 24.7% of annual rainfall) and riverbed leakage together account for only 11.4%. Recharge occurs predominantly from November to April, with no recharge from June to August. Under future scenarios, winter recharge may increase by up to 22.7%, while summer recharge could decline by up to 100%. Groundwater storage is projected to decrease by 7.3% to 58.3% compared to 2010–2020, indicating the aquifer’s vulnerability to prolonged dry periods. These findings emphasize the urgent need for adaptive water management strategies and long-term monitoring to ensure sustainable groundwater use under changing climate conditions. Full article

This study aims to develop an optimal underwriting strategy for affordable (H1 and M1) and premium (L1 and M2) electric vehicles (EVs), balancing financial risk and sustainability commitments. The research is motivated by regulatory pressures, risk management needs, and sustainability goals, necessitating an adaptation of traditional underwriting models. The study employs a modified Delphi method with industry experts to identify key risk factors, including accident risk, repair costs, battery safety, driver behavior, and PCAF carbon impact. A sensitivity analysis was conducted to examine premium adjustments under different risk scenarios, categorizing EVs into four risk segments: Low-Risk, Low-Carbon (L1); Medium-Risk, Low-Carbon (M1); Medium-Risk, High-Carbon (M2); and High-Risk, High-Carbon (H1). Findings indicate that premium EVs (L1 and M2) exhibit lower volatility in underwriting costs, benefiting from advanced safety features, lower accident rates, and reduced carbon attribution penalties. Conversely, budget EVs (H1 and M1) experience higher premium fluctuations due to greater accident risks, costly repairs, and higher carbon costs under PCAF implementation. The worst-case scenario showed a 14.5% premium increase, while the best-case scenario led to a 10.5% premium reduction. The study recommends prioritizing premium EVs for insurance coverage due to their lower underwriting risks and carbon efficiency. For budget EVs, insurers should implement selective underwriting based on safety features, driver risk profiling, and energy efficiency. Additionally, incentive-based pricing such as telematics discounts, green repair incentives, and low-carbon charging rewards can mitigate financial risks and align with net-zero insurance commitments. This research provides a structured framework for insurers to optimize EV underwriting while ensuring long-term profitability and regulatory compliance. Full article

Articulated arm coordinate measuring machines are designed for in situ use directly in manufacturing environments, enabling efficient dimensional control outside of climate-controlled laboratories. This study investigates the influence of ambient temperature variation on the accuracy of length measurements performed with the Hexagon Absolute Arm 8312. The experiment was carried out in a laboratory setting simulating typical shop floor conditions through controlled temperature changes in the range of approximately 20–31 °C. A calibrated steel gauge block was used as a reference standard, allowing separation of the influence of the measuring system from that of the measured object. The results showed that the gauge block length changed in line with the expected thermal expansion, while the articulated arm coordinate measuring machine exhibited only a minor residual thermal drift and stable performance. The experiment also revealed a constant measurement offset of approximately 22 µm, likely due to calibration deviation. As part of the study, an uncertainty budget was developed, taking into account all relevant sources of influence and enabling a more realistic estimation of accuracy under operational conditions. The study confirms that modern carbon composite articulated arm coordinate measuring machines with integrated compensation can maintain stable measurement behavior even under fluctuating temperatures in controlled environments. Full article

Non-agricultural business activity of farmers is crucial not only for stabilizing farm income but also for the multifunctional development of rural areas. Capturing changes in the level and nature of this activity supports the development of sustainable agricultural and rural policy. In this context, this study aimed to identify the scale and types of non-agricultural business activity and to recognize the main determinants of such business activities undertaken by farmers in Poland between 2002 and 2022. Sectoral-level data from the Agricultural Censuses and cyclical studies of the structure of farms and household budgets were used to approximate underlying motivations for running non-agricultural business (opportunity vs. necessity entrepreneurship). The findings indicate that, in Poland, the impact of regressive factors remains strong, pushing farmers to take on additional business activity due to the large share of small and very small farms. However, during the 21st century, a gradual spread of opportunity entrepreneurship among Polish farmers has been observed. This study highlights the rationale for supporting non-agriculture business activity motivated by progressive factors to increase the income resilience of farmer households and the sustainable development of agriculture. The article indicates the need for further research on the motives for undertaking non-agricultural economic activities by farmers and the impact of this activity on the allocation of farm resources. Full article

Technological roadmaps are essential tools for managing and planning complex projects, especially in the rapidly evolving field of space exploration. Defined as dynamic schedules, they support strategic and long-term planning while coordinating current and future objectives with particular technology solutions. Currently, the available methodologies are mostly built on experts’ opinions and in just few cases, methodologies and tools have been developed to support the decision makers with a rational approach. In any case, all the available approaches are meant to draw “ideal” maturation plans. Therefore, it is deemed essential to develop an integrate new algorithms able to decision guidelines on “non-nominal” scenarios. In this context, Politecnico di Torino, in collaboration with the European Space Agency (ESA) and Thales Alenia Space–Italia, developed the Technology Roadmapping Strategy (TRIS), a multi-step process designed to create robust and data-driven roadmaps. However, one of the main concerns with its initial implementation was that TRIS did not account for time and budget estimates specific to the space exploration environment, nor was it capable of generating alternative development paths under constrained conditions. This paper discloses two main significant updates to TRIS methodology: (1) improved time and budget estimation to better reflect the specific challenges of space exploration scenarios and (2) the capability of generating alternative roadmaps, i.e., alternative technological maturation paths in resource-constrained scenarios, balancing financial and temporal limitations. The application of the developed routines to available case studies confirms the tool’s ability to provide consistent planning outputs across multiple scenarios without exceeding 20% deviation from expert-based judgements available as reference. The results demonstrate the potential of the enhanced methodology in supporting strategic decision making in early-phase mission planning, ensuring adaptability to changing conditions, optimized use of time and financial resources, as well as guaranteeing an improved flexibility of the tool. By integrating data-driven prioritization, uncertainty modeling, and resource-constrained planning, TRIS equips mission planners with reliable tools to navigate the complexities of space exploration projects. This methodology ensures that roadmaps remain adaptable to changing conditions and optimized for real-world challenges, supporting the sustainable advancement of space exploration initiatives. Full article

Educational buildings, including schools, nurseries and universities, face stricter regulation and design control on indoor air quality (IAQ) and thermal conditions than other built environments, as these may affect children’s health and wellbeing. In this scoping review, wide-ranging health, performance, and absenteeism consequences of poor—and benefits of good—IAQ and thermal conditions are evaluated, focusing on source control, ventilation and air purification interventions. Economic impacts of interventions in educational buildings have been evaluated to enable the assessment of tangible building-related costs and savings, alongside less easily quantifiable improvements in educational attainment and reduced healthcare. Key recommendations are provided to assist decision makers in pathways to provide clean air, at an optimal temperature for students’ learning and health outcomes. Although the role of educational buildings can be challenging to isolate from other socio-economic confounders, secondary short- and long-term impacts on attainment and absenteeism have been demonstrated from the health effects associated with various pollutants. Sometimes overlooked, source control and repairing existing damage can be important cost-effective methods in minimising generation and preventing ingress of pollutants. Existing ventilation standards are often not met, even when mechanical and hybrid ventilation systems are already in place, but can often be achieved with a fraction of a typical school budget through operational and maintenance improvements, and small-scale air-cleaning and ventilation technologies, where necessary. Full article

Bird-dropping fouling and hotspot anomalies remain the most prevalent and detrimental defects in utility-scale photovoltaic (PV) plants; their co-occurrence on a single module markedly curbs energy yield and accelerates irreversible cell degradation. However, markedly disparate visual–thermal signatures of the two phenomena impede high-fidelity concurrent detection in existing robotic inspection systems, while stringent onboard compute budgets also preclude the adoption of bulky detectors. To resolve this accuracy–efficiency trade-off for dual-defect detection, we present YOLOv8-SG, a lightweight yet powerful framework engineered for mobile PV inspectors. First, a rigorously curated multi-modal dataset—RGB for stains and long-wave infrared for hotspots—is assembled to enforce robust cross-domain representation learning. Second, the HSV color space is leveraged to disentangle chromatic and luminance cues, thereby stabilizing appearance variations across sensors. Third, a single-head self-attention (SHSA) block is embedded in the backbone to harvest long-range dependencies at negligible parameter cost, while a global context (GC) module is grafted onto the detection head to amplify fine-grained semantic cues. Finally, an auxiliary bounding box refinement term is appended to the loss to hasten convergence and tighten localization. Extensive field experiments demonstrate that YOLOv8-SG attains 86.8% mAP@0.5, surpassing the vanilla YOLOv8 by 2.7 pp while trimming 12.6% of parameters (18.8 MB). Grad-CAM saliency maps corroborate that the model’s attention consistently coincides with defect regions, underscoring its interpretability. The proposed method, therefore, furnishes PV operators with a practical low-latency solution for concurrent bird-dropping and hotspot surveillance. Full article

Regional HIV-1 epidemics are evolving with distinct patterns in transmission routes, subtype distribution, and molecular transmission cluster (MTCs) characteristics. We analyzed 9500 HIV-1 cases diagnosed over 30 years using phylogenetic and network methods, integrating molecular, epidemiological, demographic, and behavioral data. Subtype A6 remains dominant nationally (80.6%), followed by 63_02A6 (7.9%), subtype B (5.6%), 02_AG_FSU (1.2%), 03_A6B (0.7%), and 14/73_BG (0.6%). Non-A6 infections were more common among males (OR 1.51) and men who have sex with men (OR 7.33). Network analysis identified 421 MTCs, with 256 active clusters. Clustering was more likely among young individuals (OR: 1.31), those not receiving antiretroviral therapy (OR: 2.70), and injecting drug users (OR: 1.28). Non-A6 subtypes showed a higher likelihood of clustering. Phylogenetic analysis revealed that local clusters of the major subtypes originated between the late 1970s (subtype B) and the mid-2000s (63_02A6) with links to populations in Eastern Europe, Central Asia (subtypes A6, 63_02A6, 02_AG_FSU, 03_A6B), and Western Europe and the Americas (subtype B, 14/73_BG). These findings indicate a complex, evolving regional epidemic transitioning from subtype A6 dominance to a more diverse mix of subtypes. The ability of non-A6 subtypes to form active MTCs suggests their establishment in the local population. Full article

Large language models (LLMs), built upon Transformer architectures, have demonstrated remarkable performance in a wide range of natural language processing tasks. However, their practical deployment—especially in long-context scenarios—is often hindered by the computational and memory costs associated with managing the key–value (KV) cache during inference. Optimizing this process is therefore crucial for improving LLM efficiency and scalability. In this study, we propose a novel entropy-guided KV caching strategy that leverages the distribution characteristics of attention scores within each Transformer layer. Specifically, we compute the entropy of attention weights for each head and use the average entropy of all heads within a layer to assess the layer’s contextual importance. Higher-entropy layers—those exhibiting broader attention dispersion—are allocated larger KV cache budgets, while lower-entropy (sink-like) layers are assigned smaller budgets. Instead of selecting different key–value tokens per head, our method selects a common set of important tokens per layer, based on aggregated attention scores, and caches them uniformly across all heads within the same layer. This design preserves the structural integrity of multi-head attention while enabling efficient token selection during the prefilling phase. The experimental results demonstrate that our approach improves cache utilization and inference speed without compromising generation quality. For example, on the Qwen3 4B model, our method reduces memory usage by 4.18% while preserving ROUGE score, and on Mistral 0.1v 7B, it reduces decoding time by 46.6%, highlighting entropy-guided layer analysis as a principled mechanism for scalable long-context language modeling. Full article

Advanced technologies, such as robotic process automation, blockchain, and machine learning, increase audit efficiency. Nonetheless, some Supreme Audit Institutions (SAIs) have not undergone digital transformation. This research aimed to develop a comprehensive framework for supreme audit institutions to adopt and integrate emerging technologies into their auditing processes using a hybrid theoretical approach based on the TAM (Technology Acceptance Model) and TOE (Technology–Organisation–Environment) models. The framework was informed by insights from nineteen highly experienced experts in the field from eight countries. Through a two-round Delphi questionnaire, the experts provided valuable input on the key factors, challenges, and strategies for successful technology adoption by public sector audit organisations. The findings of this research reveal that technology adoption in SAIs starts with solid management support led by the chief technology officer. They must evaluate the IT infrastructure and readiness for advanced technologies, considering the budget and funding. Integrating solutions like the SAI of Ghana’s Audit Management Information System can significantly enhance audit efficiency. Continuous staff training is essential to build a positive attitude toward new technologies, covering areas like data algorithm auditing and big data analysis. Assessing the complexity and compatibility of new technologies ensures ease of use and cost-effectiveness. Continuous support from technology providers and monitoring advancements will keep SAIs aligned with technological developments, enhancing their auditing capabilities. Full article

This paper proposes an incentive scheme based on a public–private partnership (PPP) to encourage low-carbon travel behavior by inducing the mode choice shift from private cars to public transit. The scheme involves three key entities: travelers, the government, and the private sector. Travelers can choose between private cars and public transit, producing different emissions. As the leader, the government aims to reduce total emission to a certain level with limited budgets. The private sector, as an intermediary, invests subsidies in low-carbon rewards to attract green travelers and benefits from a larger user pool. A two-layer multi-objective optimization model is proposed, which includes travel time, monetary cost, and emission. The objective of the upper level is to maximize the utilities of the private sector and minimize social costs to the government. The lower layer is the user equilibrium of the travelers. The numerical results obtained through heuristic algorithms demonstrate that the proposed scheme can achieve a triple-win situation, where all stakeholders benefit. Moreover, sensitivity analysis finds that prioritizing pollution control strategies will be beneficial to the government only if the unit pollution control cost coefficient is below a low threshold. Contrary to intuition, larger government subsidies do not necessarily lead to better promotion of low-carbon travel. Full article