Judith María Mendívil Gastélum, Héctor Hugo Pérez Villarreal
La presente investigación exploratoria identifica los factores que impulsan la adopción del boca a boca electrónico (eWOM) entre estudiantes universitarios y examina su relación con el amor por la marca, enfatizando cómo este vínculo emocional influye en la aceptación de reseñas en línea dentro de las redes sociales. Se aplicó un cuestionario dirigido a usuarios de Facebook para la recolección de datos, y el conjunto de datos fue analizado mediante un enfoque de modelos de ecuaciones estructurales (PLS-SEM), basado en el Modelo de Aceptación de la Información (IACM). Los hallazgos revelan que el amor por la marca desempeña un papel mediador fundamental en la relación entre la confianza en la marca y la adopción del eWOM, así como en la intención de cocreación. Los resultados indican que el apego emocional hacia la marca canaliza la confianza hacia acciones concretas de difusión y cocreación en las redes sociales. El estudio recomienda que las instituciones desarrollen campañas de marketing que fortalezcan el bienestar social y fomenten comportamientos socialmente responsables, integrando estrategias que promuevan la adopción del eWOM entre los estudiantes universitarios.
Southeast Asian coastal zone boasts a concentration of land and marine resources with vulnerable habitats, especially in highly artificial bays. Yet systematic studies on their ecological security assessment are lacking. To investigate the dynamic relationship between land use and ecological security patterns (ESP) in Langsa Bay, we assessed the ESPs for 2002–2023 and four scenarios for 2030 by satellite remote-sensing time series images. We coupled the RF-GLCM algorithm and the InVEST-PLUS model to simulate ecological services. Realizing the ESP construction and relationship dynamics analysis by circuit theory model and ecological security index (ESI) evaluation system. The approach improves the identification of ecological sources, corridors and nodes, providing an explicit spatial framework for conservation prioritization. Results showed that land-use changes strongly influenced ESP, with a 37.55 km2 increase in construction and a 159.78 km2 decrease in forest. Orchards increased by 201.79 km2, leading to a drastic decrease in forest and cropland. Future scenarios suggested that ecological and cropland protections (ESI: 4.56 and 1.98) enhance ecological security, whereas economic development (ESI: −1.20) reduces it. The findings emphasize prioritizing ecological and agricultural land preservation to mitigate ecological risks in bays, offering spatial, temporal and land-use insights for sustainable coastal management.
For precise, efficient, and safe drone landings, ground platforms should real-time, accurately locate descending drones and guide them to designated spots. While mmWave sensing combined with cameras improves localization accuracy, lower sampling frequency of traditional frame cameras compared to mmWave radar creates bottlenecks in system throughput. In this work, we upgrade traditional frame camera with event camera, a novel sensor that harmonizes in sampling frequency with mmWave radar within ground platform setup, and introduce mmE-Loc, a high-precision, low-latency ground localization system designed for precise drone landings. To fully exploit the \textit{temporal consistency} and \textit{spatial complementarity} between these two modalities, we propose two innovative modules: \textit{(i)} the Consistency-instructed Collaborative Tracking module, which further leverages the drone's physical knowledge of periodic micro-motions and structure for accurate measurements extraction, and \textit{(ii)} the Graph-informed Adaptive Joint Optimization module, which integrates drone motion information for efficient sensor fusion and drone localization. Real-world experiments conducted in landing scenarios with a drone delivery company demonstrate that mmE-Loc significantly outperforms state-of-the-art methods in both accuracy and latency.
Remote sensing imagery from systems such as Sentinel provides full coverage of the Earth's surface at around 10-meter resolution. The remote sensing community has transitioned to extensive use of deep learning models due to their high performance on benchmarks such as the UCMerced and ISPRS Vaihingen datasets. Convolutional models such as UNet and ResNet variations are commonly employed for remote sensing but typically only accept three channels, as they were developed for RGB imagery, while satellite systems provide more than ten. Recently, several transformer architectures have been proposed for remote sensing, but they have not been extensively benchmarked and are typically used on small datasets such as Salinas Valley. Meanwhile, it is becoming feasible to obtain dense spatial land-use labels for entire first-level administrative divisions of some countries. Scaling law observations suggest that substantially larger multi-spectral transformer models could provide a significant leap in remote sensing performance in these settings. In this work, we propose ChromaFormer, a family of multi-spectral transformer models, which we evaluate across orders of magnitude differences in model parameters to assess their performance and scaling effectiveness on a densely labeled imagery dataset of Flanders, Belgium, covering more than 13,500 km^2 and containing 15 classes. We propose a novel multi-spectral attention strategy and demonstrate its effectiveness through ablations. Furthermore, we show that models many orders of magnitude larger than conventional architectures, such as UNet, lead to substantial accuracy improvements: a UNet++ model with 23M parameters achieves less than 65% accuracy, while a multi-spectral transformer with 655M parameters achieves over 95% accuracy on the Biological Valuation Map of Flanders.
Semantic segmentation of SAR images has garnered significant attention in remote sensing due to the immunity of SAR sensors to cloudy weather and light conditions. Nevertheless, SAR imagery lacks detailed information and is plagued by significant speckle noise, rendering the annotation or segmentation of SAR images a formidable task. Recent efforts have resorted to annotating paired optical-SAR images to generate pseudo-labels through the utilization of an optical image segmentation network. However, these pseudo-labels are laden with noise, leading to suboptimal performance in SAR image segmentation. In this study, we introduce a more precise method for generating pseudo-labels by incorporating semi-supervised learning alongside a novel image resolution alignment augmentation. Furthermore, we introduce a symmetric cross-entropy loss to mitigate the impact of noisy pseudo-labels. Additionally, a bag of training and testing tricks is utilized to generate better land-cover mapping results. Our experiments on the GRSS data fusion contest indicate the effectiveness of the proposed method, which achieves first place. The code is available at https://github.com/StuLiu/DFC2025Track1.git.
Weixian Qian, Tianyi Yang, Sebastian Schroder
et al.
Reliable assessment of safe landing sites in unstructured environments is essential for deploying Unmanned Aerial Vehicles (UAVs) in real-world applications such as delivery, inspection, and surveillance. Existing learning-based approaches often degrade under covariate shift and offer limited transparency, making their decisions difficult to interpret and validate on resource-constrained platforms. We present NeuroSymLand, a neuro-symbolic framework for marker-free UAV landing site safety assessment that explicitly separates perception-driven world modeling from logic-based safety reasoning. A lightweight segmentation model incrementally constructs a probabilistic semantic scene graph encoding objects, attributes, and spatial relations. Symbolic safety rules, synthesized offline via large language models with human-in-the-loop refinement, are executed directly over this world model at runtime to perform white-box reasoning, producing ranked landing candidates with human-readable explanations of the underlying safety constraints. Across 72 simulated and hardware-in-the-loop landing scenarios, NeuroSymLand achieves 61 successful assessments, outperforming four competitive baselines, which achieve between 37 and 57 successes. Qualitative analysis highlights its superior interpretability and transparent reasoning, while deployment incurs negligible edge overhead. Our results suggest that combining explicit world modeling with symbolic reasoning can support accurate, interpretable, and edge-deployable safety assessment in mobile systems, as demonstrated through UAV landing site assessment.
Virginia's seventeenth- and eighteenth-century land patents survive primarily as narrative metes-and-bounds descriptions, limiting spatial analysis. This study systematically evaluates current-generation large language models (LLMs) in converting these prose abstracts into geographically accurate latitude/longitude coordinates within a focused evaluation context. A digitized corpus of 5,471 Virginia patent abstracts (1695-1732) is released, with 43 rigorously verified test cases serving as an initial, geographically focused benchmark. Six OpenAI models across three architectures-o-series, GPT-4-class, and GPT-3.5-were tested under two paradigms: direct-to-coordinate and tool-augmented chain-of-thought invoking external geocoding APIs. Results were compared against a GIS analyst baseline, Stanford NER geoparser, Mordecai-3 neural geoparser, and a county-centroid heuristic. The top single-call model, o3-2025-04-16, achieved a mean error of 23 km (median 14 km), outperforming the median LLM (37.4 km) by 37.5%, the weakest LLM (50.3 km) by 53.5%, and external baselines by 67% (GIS analyst) and 70% (Stanford NER). A five-call ensemble further reduced errors to 19.2 km (median 12.2 km) at minimal additional cost (~USD 0.20 per grant), outperforming the median LLM by 48.7%. A patentee-name redaction ablation slightly increased error (~7%), showing reliance on textual landmark and adjacency descriptions rather than memorization. The cost-effective gpt-4o-2024-08-06 model maintained a 28 km mean error at USD 1.09 per 1,000 grants, establishing a strong cost-accuracy benchmark. External geocoding tools offer no measurable benefit in this evaluation. These findings demonstrate LLMs' potential for scalable, accurate, cost-effective historical georeferencing.
Mainstreaming biodiversity into protection planning and management is of great significance for biodiversity conservation and sustainable development. Species potential distribution modeling is an effective way for species diversity evaluation and biodiversity hotspots identification, which are crucial for biodiversity conservation. Taking the Central Urban Area of Chongqing Municipality as the study area, the main objectives of this study were to identify the potentially suitable habitats, species richness and biodiversity of key protected species in current and future, determine the relative contribution of environmental factors and assess the conservation effectiveness of protected areas (PAs) based on MaxEnt model and gap analysis. The results showed that the current potentially suitable habitats of total key protected species were mainly located in “two rivers and four mountains”, with a total area of 1610.55 km2, of which forestland accounted for 59.78 %. Species suitable habitats demonstrated clear topographic heterogeneity, and the distribution index decreased at first and then increased with increasing terrain niche index (TNI). Meanwhile, it was observed that key protected plants and birds shared similar suitable habitats in mountainous areas, with an overlapping area of 753.53 km2, and the high species richness covered 182.83 km2. In 2050, the future biodiversity hotspots would remain stable and increase steadily. In terms of the direction of centroid shift, the biodiversity hotspots would migrate to low latitude, low altitude and southeast by 8.34 km. The jackknife tests indicated that the potential distribution of key protected species was mainly determined by land use, mean diurnal range and TNI. Additionally, the problems of high protection gaps and low protection effectiveness coexisted in the existing PAs, with the overlapping area of the comprehensive biodiversity hotspots and the existing PAs was only 446.96 km2. Finally, suggestions for natural PAs system optimization and ecological protection were proposed. This study provides scientific supports for biodiversity conservation and efficient management.
With the advantages of high efficiency and convenience, shared bikes are rapidly gaining popularity in China, and become an important connection mode for rail transit stations, but it also brings many problems which affect urban transportation and environment. Based on multi-source big data such as Xiamen shared bikes and urban space, this article analyzes the spatial and temporal characteristics of shared bikes travel by Python and GIS, and explores their riding characteristics and the effects concerning land use, urban environment and other aspects in the station area. Shared bikes have the characteristics of short spatial and temporal distance utilization, which provides an important support for the connecting traffic of urban subway stations. The connection and utilization of shared bikes around the station are affected by various urban factors, which also reflect the stages and differences of the station development. By analyzing the characteristics of shared bikes trips, rail transit stations can be classified into four types, and the develop strategies should be taken according to the difference of station types, to improve the urban slow traffic system, promote the development level of rail transit stations.(共享单车凭借其高效便利的优势在全国各大城市快速普及，已成为轨道交通站点重要的接驳方式，但其供需失衡、乱停乱放等问题也给城市交通与环境带来挑战。利用厦门市共享单车及城市空间等多源大数据，通过Python、GIS解析共享单车出行的时空特征，探究轨道站域骑行特征及其来自土地利用、城市建成环境等方面的影响效应，结果表明：共享单车具有短时空距离利用的特征，为城市地铁站点的接驳交通提供了重要支撑；站点周边共享单车的接驳利用受城市多种因素影响，同时也反映了站点开发的阶段性和差异性；根据共享单车出行特征，轨道交通站点可归纳为四类，应根据站点的不同类型因站施策，完善城市慢行系统，促进共享单车骑行，加强轨道交通站域开发。)
Synthetic Aperture Radar (SAR) images have proven to be a valuable cue for multimodal Land Cover Classification (LCC) when combined with RGB images. Most existing studies on cross-modal fusion assume that consistent feature information is necessary between the two modalities, and as a result, they construct networks without adequately addressing the unique characteristics of each modality. In this paper, we propose a novel architecture, named the Asymmetric Semantic Aligning Network (ASANet), which introduces asymmetry at the feature level to address the issue that multi-modal architectures frequently fail to fully utilize complementary features. The core of this network is the Semantic Focusing Module (SFM), which explicitly calculates differential weights for each modality to account for the modality-specific features. Furthermore, ASANet incorporates a Cascade Fusion Module (CFM), which delves deeper into channel and spatial representations to efficiently select features from the two modalities for fusion. Through the collaborative effort of these two modules, the proposed ASANet effectively learns feature correlations between the two modalities and eliminates noise caused by feature differences. Comprehensive experiments demonstrate that ASANet achieves excellent performance on three multimodal datasets. Additionally, we have established a new RGB-SAR multimodal dataset, on which our ASANet outperforms other mainstream methods with improvements ranging from 1.21% to 17.69%. The ASANet runs at 48.7 frames per second (FPS) when the input image is 256x256 pixels. The source code are available at https://github.com/whu-pzhang/ASANet
Do ecosystems primarily reflect evolutionary history or current environment? Predicting land-atmosphere exchange hinges on this unresolved question. Plant traits adapt to particular environments over evolutionary timescales, yet their individual relationships with current climate and soils are often obscured by limited sampling, plant-type effects, and multiple adaptive strategies that can yield similar outcomes. Crucially, it is the coordination of traits, rather than any single trait, that governs vegetation dynamics and ecosystem fluxes, yet such multivariate relationships cannot be directly observed. We present DifferLand, a differentiable hybrid model that integrates process understanding with machine learning to uncover latent trait-environment relationships from global satellite and in-situ observations (2001-2023). DifferLand explains up to 88% of the variance in canopy structure, photosynthesis, and carbon exchange by learning latent ecological axes-leaf economics, plant stature, and cropland distribution-that link long-term adaptation with short-term dynamics. Interpretable machine learning shows that these coordinated axes emerge from nonlinear interactions between plant-type attributes and local environment. Embedding such relationships into terrestrial models establishes a pathway toward adaptive models that better predict ecosystem resilience under climate change.
Dilcy Morgana Barros Maciel Cabral Davino, Ianara Acioli de Freitas Melo, Samara Almeida de Souza Griz
et al.
Introdução: A automedicação, prática rotineira vivenciada por grande parte da população, consiste no consumo de um determinado medicamento sem prescrição médica, para aliviar os sintomas e tratar doenças. Quando praticada corretamente, a automedicação pode também contribuir para aliviar financeiramente os sistemas de saúde pública. Porém, o uso indiscriminado dos medicamentos, pode acarretar resultados indesejáveis. A informação adquirida sobre o uso de medicamentos proporciona maior confiança para a prática entre os estudantes universitários, em especial, os da área da saúde. No Brasil, mais pesquisas e ações precisam ser realizadas, pois há deficiência de dados úteis para combater a automedicação irresponsável, estimulando o uso racional de medic amentos conforme recomendado pela OMS. Objetivo: Avaliar o perfil de consumo de medicamentos por graduandos do curso de farmácia, além de esclarecer e conscientizar os futuros profissionais farmacêuticos sobre automedicação, pois serão os principais responsáveis por orientar a população sobre os riscos dessa prática. Método: Foi realizado um estudo transversal de caráter descritivo com abordagem quantitativa, com acadêmicos matriculados no 1o semestre de 2018 em todos os períodos do curso de farmácia de uma instituição de ensino superior, no município de Maceió-AL. Os dados foram coletados, após a aprovação do Comitê de Ética em Pesquisa (CEP) divulgada através do Parecer Consubstanciado (número: 2.436.431), através da aplicação de um questionário semi-estruturado e assinatura do Termo de Consentimento Livre e Esclarecido (TCLE) pelos alunos concordaram em participar da pesquisa. Resultados: O estudo contou com amostra de 225 graduandos, sendo 74,2% do sexo feminino, prevalecendo a faixa etária de 21 a 30 anos e o estado civil solteiro. Apesar de 88,4% relatar ter conhecimento sobre a automedicação, os resultados indicaram a prática por parte dos discentes, pois 68,4% informou se automedicar, 58,2% adquirem medicamentos sem receita, 86,2% estocam medicamentos em casa, as principais classes consumidas são os analgésicos, na forma farmacêutica de comprimidos e o principal motivo foi para dores em geral, tanto para automedicação como de uso contínuo. Quanto à influência no uso de medicamentos, 56,0% consultam o farmacêutico, 32,9% receberam indicações por parte de familiares, amigos ou vizinhos e 53,3% as propagandas de medicamentos chamam atenção para compra, principalmente por TV/Rádio. Quando analisada a mudança de no perfil de consumo dos graduandos, notou-se que 72,30% mudou seu comportamento, e destes, 84,4% reduziram o consumo de medicamentos. Conclusão: Felizmente, foi verificado que existe uma mudança positiva no comportamento de consumo com redução em todos os períodos, provavelmente relacionada ao conhecimento adquirido durante a graduação, tendo consciência dos danos que a automedicação inconsciente pode causar à saúde.
Pharmacy and materia medica, Pharmaceutical industry
Evapotranspiration (ET) is a critical concern for water management and hydrological cycle; thus, studies of ET have been performed to aid irrigation and water resource planning. Moreover, global warming-related studies are critical, as sensible heat contributes to warming, while the latent heat flux contributes to cooling. Recently, FLUXNET2015, a large energy flux dataset comprising climatic elements, was updated with a corrected heat balance relationship. In this study, we aim to applicability of the inverse analysis (IA) for estimating farmland ET. Practically, we evaluated the estimated ET (LEest) consistency using IA, which compared common climate data with observed data (LEobs) from US-Ne1 (irrigated), US-Ne2 (irrigated), and US-Ne3 (non-irrigated) land in FLUXNET2015. For an hourly time step, net radiation (Rn) and heat flux into the ground (G) were reasonably allocated into sensible (H) and latent (LE) heat fluxes, and LEobs was reasonably reproduced by LEest. For daily and monthly time steps, LEobs was reproduced well by LEest, with similar accuracies. For a yearly time step, LEobs was reproduced by LEest with an R2 of 0.933. Reasonability of the IA method also confirmed ET in crop growing season by comparing LEobs and LEest. A cooling effect under the canopy was observed on irrigated farmland in eight of the 22 analyzed years, whereas non-irrigated farmland did not exhibit a cooling effect. The maximum cooling effect was 4.26 °C of the monthly average. The results confirm that IA can be applied to non-irrigated and irrigated farmland if a cooling effect is not observed. IA can therefore be used to improve farmland water utilization because of accurate LEest and determining the capacities of irrigation facilities. The findings can be used to evaluate cooling effects on farmland, as well as reasonable allocations of Rn into H and LE, which promote the advancement of global warming issues.
This paper provides theoretical foundations for the existence of national social innovations systems (NSIS) and presents such a system with empirical data. Departing from the activities in France of Ashoka, a large and old service organization, which we label as knowledge-intensive social service (KISS), we build a large and robust social innovation network in France and argue that it represents a credible approximation of the country’s NSIS. On this basis, we find differences within the national innovation system (NIS). Indeed, the core of the NSIS involves very few actors emanating from manufacturing or technology-intensive industries, and the coordination between actors seems more bottom-up than in the NIS.
We present a framework that uses GAN-augmented images to complement certain specific attributes, usually underrepresented, for machine learning model training. This allows us to improve inference quality over those attributes for the facial recognition tasks.
Land-cover classification has long been a hot and difficult challenge in remote sensing community. With massive High-resolution Remote Sensing (HRS) images available, manually and automatically designed Convolutional Neural Networks (CNNs) have already shown their great latent capacity on HRS land-cover classification in recent years. Especially, the former can achieve better performance while the latter is able to generate lightweight architecture. Unfortunately, they both have shortcomings. On the one hand, because manual CNNs are almost proposed for natural image processing, it becomes very redundant and inefficient to process HRS images. On the other hand, nascent Neural Architecture Search (NAS) techniques for dense prediction tasks are mainly based on encoder-decoder architecture, and just focus on the automatic design of the encoder, which makes it still difficult to recover the refined mapping when confronting complicated HRS scenes. To overcome their defects and tackle the HRS land-cover classification problems better, we propose AutoLC which combines the advantages of two methods. First, we devise a hierarchical search space and gain the lightweight encoder underlying gradient-based search strategy. Second, we meticulously design a lightweight but top-performing decoder that is adaptive to the searched encoder of itself. Finally, experimental results on the LoveDA land-cover dataset demonstrate that our AutoLC method outperforms the state-of-art manual and automatic methods with much less computational consumption.
Lukas Höhndorf, Thomas Nagler, Phillip Koppitz
et al.
The RTS smoother is widely used for state estimation and it is utilized here to increase the data quality with respect to physical coherence and to increase resolution. The purpose of this paper is to enhance the performance of the RTS smoother to reconstruct an aircraft landing using on board recorded data only. Thereby, errors and uncertainties of operational flight data (e.g. altitude, attitude, position, speed) recorded during flights of civil aircraft are minimized. These data can be used for subsequent analyses in terms of flight safety or efficiency, which is commonly referred to as Flight Data Monitoring (FDM). Statistical assumptions of the smoother theory are not always verified during application but (consciously or not) assumed to be fulfilled. These assumptions can hardly be verified prior to the smoother application, however, they can be verified using the results of an initial smoother iteration and modifications of specific smoother characteristics can be suggested. This project specifically verifies assumptions on the measurement noise characteristics. Variance and covariance of the measurement noise can be checked after the initial smoother application. It is discovered that these characteristics change over time and should be accounted for with a time varying covariance matrix. This sequence of matrices is estimated by kernel smoothing and replaces an initially assumed fixed and diagonal covariance matrix used for the first smoother run. The results of this second smoother iteration are mostly improved compared to the initial iteration, i.e. the errors are significantly reduced. Subsequently, the remaining dependence structures of the residuals of the second smoother iteration can be captured by copula models. Their interpretation is useful for a revision of the physical model utilized by the RTS smoother.