Search Results (2,602)

Ensuring software quality during development requires effective regression testing. However, test suites in open-source libraries often grow large, redundant, and difficult to maintain. Most traditional test suite optimization methods treat test cases as atomic units, without analyzing the utility of individual instructions. This paper presents an intelligent method for test case synthesis using a Q-learning agent. The agent learns to construct compact test cases by interacting with an execution environment and receives rewards based on branch coverage improvements and simultaneous reductions in test case length. The training process includes a pretraining phase that transfers knowledge from the original test suite, followed by adaptive learning episodes on individual test cases. As a result, the method requires no formal documentation or API specifications and uses only execution traces of the original test cases. An explicit synthesis algorithm constructs new test cases by selecting API calls from a learned policy encoded in a Q-table. Experiments were conducted on two open-source C++ libraries of differing API complexity and original test suite size. The results show that the proposed method can reach up to 67% test suite reduction while preserving branch coverage, confirming its effectiveness for regression test suite minimization in resource-constrained or specification-limited environments. Full article

The implementation of the trans-European tourist cycle route network “EuroVelo” is fostering new strategic importance for non-motorized mobility and the associated practice of cycling tourism. Indeed, slow tourism offers a pathway for the development of inland areas. The infrastructure supporting it, such as long-distance cycling and walking paths, can act as a vital connection, stimulating regeneration in peripheral territories by enhancing environmental and landscape assets, as well as preserving heritage, local identity, and culture. The regeneration of peri-urban landscapes through soft mobility is recognized as the cornerstone for accessibility to material and immaterial resources (including ecosystem services) for multiple categories of users, including the most vulnerable, especially following the restoration of green-area systems and non-urbanized areas with degraded ecosystems. Considering the forthcoming implementation of the Magna Grecia cycling route, the southernmost segment of the “EuroVelo” network traversing three regions in southern Italy, this contribution briefly examines the necessity of defining new development policies to effectively integrate sustainable slow tourism with the enhancement of environmental and landscape values in the coastal areas along the route. Specifically, this case study focuses on a coastal stretch characterized by significant morphological and environmental features and notable landscapes interwoven with densely built environments. In this area, environmental and landscape values face considerable threats from scattered, irregular, low-density settlements, abandoned sites, and other inappropriate constructions along the coastline. Full article

Deep marine shale is the primary carrier of shale gas resources in Southwestern China. Because the occurrence and gas content of methane vary with burial conditions, understanding the microscopic mechanism of methane occurrence in deep marine shale is critical for effective shale gas exploitation. The temperature and pressure conditions in deep shale exceed the operating limits of experimental equipment; thus, few studies have discussed the microscopic occurrence mechanism of shale gas in deep marine shale. This study applies molecular simulation technology to reveal the methane’s microscopic occurrence mechanism, particularly the main controlling factor of adsorbed methane in deep marine shale. Two types of simulation models are also proposed. The Grand Canonical Monte Carlo (GCMC) method is used to simulate the adsorption behavior of methane molecules in these two models. The results indicate that the isosteric adsorption heat of methane in both models is below 42 kJ/mol, suggesting that methane adsorption in deep shale is physical adsorption. Adsorbed methane concentrates on the pore wall surface and forms a double-layer adsorption. Furthermore, adsorbed methane can transition to single-layer adsorption if the pore size is less than 1.6 nm. The total adsorption capacity increases with rising pressure, although the growth rate decreases. Excess adsorption capacity is highly sensitive to pressure and can become negative at high pressures. Methane adsorption capacity is determined by pore size and adsorption potential, while accommodation space and adsorption potential are influenced by pore size and mineral type. Under deep marine shale reservoir burial conditions, with burial depth deepening, the effect of temperature on shale gas occurrence is weaker than pressure. Higher temperatures inhibit shale gas occurrence, and high pressure enhances shale gas preservation. Smaller pores facilitate the occurrence of adsorbed methane, and larger pores have larger total methane adsorption capacity. Deep marine shale with high formation pressure and high clay mineral content is conducive to the microscopic accumulation of shale gas in deep marine shale reservoirs. This study discusses the microscopic occurrence state of deep marine shale gas and provides a reference for the exploration and development of deep shale gas. Full article

Recent archaeological discoveries across the Aegean, Cyprus, and western Anatolia have renewed interest in pre-Neolithic seafaring and early island colonization. However, the environmental contexts that support such early coastal occupations remain poorly understood, largely due to the submergence of Pleistocene shorelines following post-glacial sea-level rise. This study addresses this gap through an integrated geoarchaeological investigation of the pre-Neolithic site of Ouriakos on Lemnos Island, northeastern Aegean (Greece), dated to the mid-11th millennium BCE. By reconstructing both the terrestrial and submerged paleolandscapes of the site, we examine ecological conditions, resource availability, and sedimentary processes that shaped human activity and site preservation. Employing a multiscale methodological approach—combining bathymetric survey, geomorphological mapping, soil micromorphology, geochemical analysis, and Optically Stimulated Luminescence (OSL) dating—we present a comprehensive framework for identifying and interpreting early coastal settlements. Stratigraphic evidence reveals phases of fluvial, aeolian, and colluvial deposition associated with an alternating coastline. The core findings reveal that Ouriakos was established during a phase of environmental stability marked by paleosol development, indicating sustained human presence. By bridging terrestrial and marine data, this research contributes significantly to the understanding of human coastal mobility during the Pleistocene–Holocene transition. Full article

Malware evolution presents growing security threats for resource-constrained Internet of Medical Things (IoMT) devices. Conventional federated learning (FL) often suffers from slow convergence, high communication overhead, and fairness issues in dynamic IoMT environments. In this paper, we propose FedGCL, a secure and efficient FL framework integrating contrastive graph representation learning for enhanced feature discrimination, a Jain-index-based fairness-aware aggregation mechanism, an adaptive synchronization scheduler to optimize communication rounds, and secure aggregation via homomorphic encryption within a Trusted Execution Environment. We evaluate FedGCL on four benchmark malware datasets (Drebin, Malgenome, Kronodroid, and TUANDROMD) using 5 to 15 graph neural network clients over 20 communication rounds. Our experiments demonstrate that FedGCL achieves 96.3% global accuracy within three rounds and converges to 98.9% by round twenty—reducing required training rounds by 45% compared to FedAvg—while incurring only approximately 10% additional computational overhead. By preserving patient data privacy at the edge, FedGCL enhances system resilience without sacrificing model performance. These results indicate FedGCL’s promise as a secure, efficient, and fair federated malware detection solution for IoMT ecosystems. Full article

Multi-modal object detection that fuses RGB (Red-Green-Blue) and infrared (IR) data has emerged as an effective approach for addressing challenging visual conditions such as low illumination, occlusion, and adverse weather. However, most existing multi-modal detectors prioritize accuracy while neglecting computational efficiency, making them unsuitable for deployment on resource-constrained edge devices. To address this limitation, we propose PONet, a lightweight and efficient multi-modal vehicle detection network tailored for real-time edge inference. PONet incorporates Polarized Self-Attention to improve feature adaptability and representation with minimal computational overhead. In addition, a novel fusion module is introduced to effectively integrate RGB and IR modalities while preserving efficiency. Experimental results on the VEDAI dataset demonstrate that PONet achieves a competitive detection accuracy of 82.2% mAP@0.5 while sustaining a throughput of 34 FPS on the OrangePi AIpro 20T device. With only 3.76 M parameters and 10.2 GFLOPs (Giga Floating Point Operations), PONet offers a practical solution for edge-oriented remote sensing applications requiring a balance between detection precision and computational cost. Full article

This study presents a lightweight autoencoder-based approach for anomaly detection in digit recognition using federated learning on resource-constrained embedded devices. We implement and evaluate compact autoencoder models on the ESP32-CAM microcontroller, enabling both training and inference directly on the device using 32-bit floating-point arithmetic. The system is trained on a reduced MNIST dataset (1000 resized samples) and evaluated using EMNIST and MNIST-C for anomaly detection. Seven fully connected autoencoder architectures are first evaluated on a PC to explore the impact of model size and batch size on training time and anomaly detection performance. Selected models are then re-implemented in the C programming language and deployed on a single ESP32 device, achieving training times as short as 12 min, inference latency as low as 9 ms, and F1 scores of up to 0.87. Autoencoders are further tested on ten devices in a real-world federated learning experiment using Wi-Fi. We explore non-IID and IID data distribution scenarios: (1) digit-specialized devices and (2) partitioned datasets with varying content and anomaly types. The results show that small unmodified autoencoder models can be effectively trained and evaluated directly on low-power hardware. The best models achieve F1 scores of up to 0.87 in the standard IID setting and 0.86 in the extreme non-IID setting. Despite some clients being trained on corrupted datasets, federated aggregation proves resilient, maintaining high overall performance. The resource analysis shows that more than half of the models and all the training-related allocations fit entirely in internal RAM. These findings confirm the feasibility of local float32 training and collaborative anomaly detection on low-cost hardware, supporting scalable and privacy-preserving edge intelligence. Full article

Many hospitals still lack digital traceability in hygiene and cleaning management, leading to operational inefficiencies and inconsistent quality control. This study aims to establish cleaning and hygiene processes in healthcare services that are planned in accordance with standards, as well as to enhance the traceability and sustainability of these processes through digitalization. This study proposes a Hyperledger Fabric-based blockchain architecture to establish a reliable and transparent quality assurance system in process management. The proposed Quality Assurance Model utilizes digital technologies and IoT-based RFID devices to ensure the transparent and reliable monitoring of cleaning processes. Operational data related to cleaning processes are automatically recorded and secured using a decentralized blockchain infrastructure. The permissioned nature of Hyperledger Fabric provides a more secure solution compared to traditional data management systems in the healthcare sector while preserving data privacy. Additionally, the execute–order–validate mechanism supports effective data sharing among stakeholders, and consensus algorithms along with chaincode rules enhance the reliability of processes. A working prototype was implemented and validated using Hyperledger Caliper under resource-constrained cloud environments, confirming the system’s feasibility through over 100 TPS throughput and zero transaction failures. Through the proposed system, cleaning/hygiene processes in patient rooms are conducted securely, contributing to the improvement of quality standards in healthcare services. Full article

Accurate and efficient cloud detection is critical for maintaining the usability of optical remote sensing imagery, particularly in large-scale Earth observation systems. In this study, we propose FEMNet, a lightweight dual-branch network that combines state space modeling with convolutional encoding for multi-class cloud segmentation. The Mamba-based encoder captures long-range semantic dependencies with linear complexity, while a parallel CNN path preserves spatial detail. To address the semantic inconsistency across feature hierarchies and limited context perception in decoding, we introduce the following two targeted modules: a cross-stage semantic enhancement (CSSE) block that adaptively aligns low- and high-level features, and a multi-scale context aggregation (MSCA) block that integrates contextual cues at multiple resolutions. Extensive experiments on five benchmark datasets demonstrate that FEMNet achieves state-of-the-art performance across both binary and multi-class settings, while requiring only 4.4M parameters and 1.3G multiply–accumulate operations. These results highlight FEMNet’s suitability for resource-efficient deployment in real-world remote sensing applications. Full article

Dark skies, a vital natural and cultural resource, have been increasingly threatened by light pollution due to rapid urbanization, leading to ecological degradation and biodiversity loss. As a key strategy for sustainable regional development, dark sky parks (DSPs) not only preserve nocturnal environments but also enhance livability by balancing urban expansion and ecological conservation. This study develops a novel framework for evaluating DSP suitability, integrating ecological and socio-economic dimensions, including the resource base (e.g., nighttime light levels, meteorological conditions, and air quality) and development conditions (e.g., population density, transportation accessibility, and tourism infrastructure). Using the Guangdong–Hong Kong–Macao Greater Bay Area (GBA) as a case study, we employ Delphi expert consultation, GIS spatial analysis, and multi-criteria decision-making to identify optimal DSP locations and prioritize conservation zones. Our key findings reveal the following: (1) spatial heterogeneity in suitability, with high-potential zones being concentrated in the GBA’s northeastern, central–western, and southern regions; (2) ecosystem advantages of forests, wetlands, and high-elevation areas for minimizing light pollution; (3) coastal and island regions as ideal DSP sites due to the low light interference and high ecotourism potential. By bridging environmental assessments and spatial planning, this study provides a replicable model for DSP site selection, offering policymakers actionable insights to integrate dark sky preservation into sustainable urban–regional development strategies. Our results underscore the importance of DSPs in fostering ecological resilience, nighttime tourism, and regional livability, contributing to the broader discourse on sustainable landscape planning in high-urbanization contexts. Full article

Heritage Building Information Modeling (HBIM) has emerged as a key tool in advancing heritage conservation and sustainable management. Preceding reviews had typically concentrated on specific technical aspects but did not provide sufficient bibliometric analysis. This study aims to integrate existing HBIM research to identify key research patterns, emerging trends, and forecast future directions. A total of 1516 documents were initially retrieved from the Web of Science Core Collection using targeted search terms. Following a relevance screening, 1175 documents were related to the topic. CiteSpace 6.4.R1, VOSviewer 1.6.20, and Bibliometrix 4.1, three bibliometric tools, were employed to conduct both quantitative and qualitative assessments. The results show three historical phases of HBIM, identify core journals, influential authors, and leading regions, and extract six major keyword clusters: risk assessment, data acquisition, semantic annotation, digital twins, and energy and equipment management. Nine co-citation clusters further outline the foundational literature in the field. The results highlight growing scholarly interest in workflow integration and digital twin applications. Future projections emphasize the transformative potential of artificial intelligence in HBIM, while also recognizing critical implementation barriers, particularly in developing countries and resource-constrained contexts. This study provides a comprehensive and systematic framework for HBIM research, offering valuable insights for scholars, practitioners, and policymakers involved in heritage preservation and digital management. Full article

Obesity is a growing global health concern with widespread impacts on physical, psychological, and social well-being. Clinically, it is a major driver of type 2 diabetes (T2D), cardiovascular disease (CVD), non-alcoholic fatty liver disease (NAFLD), and cancer, reducing life expectancy by 5–20 years and imposing a staggering economic burden of USD 2 trillion annually (2.8% of global GDP). Despite its significant health and socioeconomic impact, earlier obesity medications, such as fenfluramine, sibutramine, and orlistat, fell short of expectations due to limited effectiveness, serious side effects including valvular heart disease and gastrointestinal issues, and high rates of treatment discontinuation. The advent of glucagon-like peptide-1 (GLP-1) receptor agonists (e.g., semaglutide, tirzepatide) has revolutionized obesity management. These agents demonstrate unprecedented efficacy, achieving 15–25% mean weight loss in clinical trials, alongside reducing major adverse cardiovascular events by 20% and T2D incidence by 72%. Emerging therapies, including oral GLP-1 agonists and triple-receptor agonists (e.g., retatrutide), promise enhanced tolerability and muscle preservation, potentially bridging the efficacy gap with bariatric surgery. However, challenges persist. High costs, supply shortages, and unequal access pose significant barriers to the widespread implementation of obesity treatment, particularly in low-resource settings. Gastrointestinal side effects and long-term safety concerns require close monitoring, while weight regain after medication discontinuation emphasizes the need for ongoing adherence and lifestyle support. This review highlights the transformative potential of incretin-based therapies while advocating for policy reforms to address cost barriers, equitable access, and preventive strategies. Future research must prioritize long-term cardiovascular outcome trials and mitigate emerging risks, such as sarcopenia and joint degeneration. A multidisciplinary approach combining pharmacotherapy, behavioral interventions, and systemic policy changes is critical to curbing the obesity epidemic and its downstream consequences. Full article

The ongoing genetic erosion of natural Prunus armeniaca populations in their native habitats underscores the urgent need for targeted conservation and restoration strategies. This study provides the first comprehensive characterization of P. armeniaca populations in the Almaty region of Kazakhstan, integrating morphological descriptors (46 parameters), molecular markers, geobotanical, and remote sensing analyses. Geobotanical and remote sensing analyses enhanced understanding of accession distribution, geological features, and ecosystem health across sites, while also revealing their vulnerability to various biotic and abiotic threats. Of 111 morphologically classified accessions, 54 were analyzed with 13 simple sequence repeat (SSR) markers and four DNA barcoding regions. Our findings demonstrate the necessity of integrated morphological and molecular analyses to differentiate closely related accessions. Genetic analysis identified 11 distinct populations with high heterozygosity and substantial genetic variability. Eight populations exhibited 100% polymorphism, indicating their potential as sources of adaptive genetic diversity. Cluster analysis grouped populations into three geographic clusters, suggesting limited gene flow across Gorges (features of a mountainous landscape) and greater connectivity within them. These findings underscore the need for site-specific conservation strategies, especially for genetically distinct, isolated populations with unique allelic profiles. This study provides a valuable foundation for prioritizing conservation targets, confirming genetic redundancies, and preserving genetic uniqueness to enhance the efficiency and effectiveness of the future conservation and use of P. armeniaca genetic resources in the region. Full article

The integration of unmanned aerial vehicles (UAVs) into next-generation wireless networks greatly enhances the flexibility and efficiency of communication and distributed computation for ground mobile devices. Federated learning (FL) provides a privacy-preserving paradigm for device collaboration but remains highly vulnerable to poisoning attacks and is further challenged by the resource constraints and heterogeneous data common to UAV-assisted systems. Existing robust aggregation and anomaly detection methods often degrade in efficiency and reliability under these realistic adversarial and non-IID settings. To bridge these gaps, we propose FedULite, a lightweight and robust federated learning framework specifically designed for UAV-assisted environments. FedULite features unsupervised local representation learning optimized for unlabeled, non-IID data. Moreover, FedULite leverages a robust, adaptive server-side aggregation strategy that uses cosine similarity-based update filtering and dimension-wise adaptive learning rates to neutralize sophisticated data and model poisoning attacks. Extensive experiments across diverse datasets and adversarial scenarios demonstrate that FedULite reduces the attack success rate (ASR) from over 90% in undefended scenarios to below 5%, while maintaining the main task accuracy loss within 2%. Moreover, it introduces negligible computational overhead compared to standard FedAvg, with approximately 7% additional training time. Full article

The report examines food availability and demand in the Anthropocene era, exploring the connections between global population growth and carrying capacity through an extended version of Cohen’s Condorcet concept. It recalls the super-Malthus and Verhulst-type scalings, matched with the recently introduced analytic relative growth rate. It focuses particularly on the ongoing Fifth Industrial Revolution (IR) and its interaction with the concept of a sustainable civilization. In this context, the significance of innovative food preservation technologies that can yield high-quality foods with health-promoting features, while simultaneously increasing food quantities and reducing adverse environmental impacts, is discussed. To achieve this, high-pressure preservation and processing (HPP) can play a dominant role. High-pressure ‘cold pasteurization’, related to room-temperature processing, has already achieved a global scale. Its superior features are notable and are fairly correlated with social expectations of a sustainable society and the technological tasks of the Fifth Industrial Revolution. The discussion is based on the authors’ experiences in HPP-related research and applications. The next breakthrough could be HPP-related sterilization. The innovative HPP path, supported by the colossal barocaloric effect, is presented. The mass implementation of pressure-related sterilization could lead to milestone societal, pro-health, environmental, and economic benefits. Full article