Search Results (18,743)

The phase-sensitive optical time-domain reflectometry (Φ-OTDR) system has shown substantial potential in distributed acoustic sensing applications. Accurate event classification is crucial for effective deployment of Φ-OTDR systems, and various methods have been proposed for event classification in Φ-OTDR systems. However, most existing methods typically rely on sufficient labeled signal data for model training, which poses a major bottleneck in applying these methods due to the expensive and laborious process of labeling extensive data. To address this limitation, we propose CLWTNet, a novel contrastive representation learning method enhanced with wavelet transform convolution for event classification in Φ-OTDR systems. CLWTNet learns robust and discriminative representations directly from unlabeled signal data by transforming time-domain signals into STFT images and employing contrastive learning to maximize inter-class separation while preserving intra-class similarity. Furthermore, CLWTNet incorporates wavelet transform convolution to enhance its capacity to capture intricate features of event signals. The experimental results demonstrate that CLWTNet achieves competitive performance with the supervised representation learning methods and superior performance to unsupervised representation learning methods, even when training with unlabeled signal data. These findings highlight the effectiveness of CLWTNet in extracting discriminative representations without relying on labeled data, thereby enhancing data efficiency and reducing the costs and effort involved in extensive data labeling in practical Φ-OTDR system applications. Full article

Soft data, such as seismic imagery, plays a critical role in subsurface modeling by providing indirect constraints away from hard data locations. However, validating whether subsurface model realizations honor this type of data remains a challenge due to the lack of robust quantitative tools. This study introduces a machine learning-based workflow for soft data checking that uses an autoencoder (AE) to encode 2D seismic slices into a latent space. Subsurface model realizations are transformed into the same domain and projected into this latent space, enabling both visual and quantitative comparisons using principal component analysis and Euclidean distances. We demonstrate the workflow on rule-based models and their associated synthetic seismic data (soft data), showing that models with similar Markov chain parameters to the reference soft data score higher in proximity metrics. This approach provides a scalable, quantitative, and interpretable framework for evaluating the consistency between soft data and subsurface models, supporting better decision-making in reservoir characterization and other geoscience applications. Full article

Previous reviews have examined specific facets of Geographic Information Systems (GIS) in transportation planning, such as transit-focused applications and open source geospatial tools. However, this study offers the first systematic, PRISMA-guided longitudinal evaluation of GIS integration in transportation planning, spanning thematic domains, data models, methodologies, and outcomes from 2004 to 2024. This study addresses this gap through a longitudinal analysis of GIS-based transportation research from 2004 to 2024, adhering to PRISMA guidelines. By conducting a mixed-methods analysis of 241 peer-reviewed articles, this study delineates major trends, such as increased emphasis on sustainability, equity, stakeholder involvement, and the incorporation of advanced technologies. Prominent domains include land use–transportation coordination, accessibility, artificial intelligence, real-time monitoring, and policy evaluation. Expanded data sources, such as real-time sensor feeds and 3D models, alongside sophisticated modeling techniques, enable evidence-based, multifaceted decision-making. However, challenges like data limitations, ethical concerns, and the need for specialized expertise persist, particularly in developing regions. Future geospatial innovations should prioritize the responsible adoption of emerging technologies, inclusive capacity building, and environmental justice to foster equitable and efficient transportation systems. This review highlights GIS’s evolution from a supplementary tool to a cornerstone of data-driven, sustainable urban mobility planning, offering insights for researchers, practitioners, and policymakers to advance transportation strategies that align with equity and sustainability goals. Full article

In the era of smart agriculture, accurate identification of plant species is critical for effective crop management, biodiversity monitoring, and the sustainable use of medicinal resources. However, existing deep learning approaches often underperform when applied to fine-grained plant classification tasks due to the lack of domain-specific, high-quality datasets and the limited representational capacity of traditional architectures. This study addresses these challenges by introducing a novel, well-curated leaf image dataset consisting of 39 classes of medicinal and aromatic plants collected from the Aromatic and Medicinal Plant Research Station in Odakkali, Kerala, India. To overcome performance bottlenecks observed with a baseline Convolutional Neural Network (CNN) that achieved only 44.94% accuracy, we progressively enhanced model performance through a series of architectural innovations. These included the use of a pre-trained VGG16 network, data augmentation techniques, and fine-tuning of deeper convolutional layers, followed by the integration of Squeeze-and-Excitation (SE) attention blocks. Ultimately, we propose a hybrid deep learning architecture that combines VGG16 with Batch Normalization, Gated Recurrent Units (GRUs), Transformer modules, and Dilated Convolutions. This final model achieved a peak validation accuracy of 95.24%, significantly outperforming several baseline models, such as custom CNN (44.94%), VGG-19 (59.49%), VGG-16 before augmentation (71.52%), Xception (85.44%), Inception v3 (87.97%), VGG-16 after data augumentation (89.24%), VGG-16 after fine-tuning (90.51%), MobileNetV2 (93.67), and VGG16 with SE block (94.94%). These results demonstrate superior capability in capturing both local textures and global morphological features. The proposed solution not only advances the state of the art in plant classification but also contributes a valuable dataset to the research community. Its real-world applicability spans field-based plant identification, biodiversity conservation, and precision agriculture, offering a scalable tool for automated plant recognition in complex ecological and agricultural environments. Full article

Rotating compressors are foundational in various industrial processes, particularly in the oil-and-gas sector, where reliable fault detection is crucial for maintaining operational continuity. While Graph Attention Network (GAT) frameworks are widely available, this study advances the state of the art by introducing a Bayesian GAT method specifically tailored for vibration-based compressor fault diagnosis. The approach integrates domain-specific digital-twin simulations built with Rotordynamic software (1.3.0), and constructs dual adjacency matrices to encode both physically informed and data-driven sensor relationships. Additionally, a hybrid forecasting-and-reconstruction objective enables the model to capture short-term deviations as well as long-term waveform fidelity. Monte Carlo dropout further decomposes prediction uncertainty into aleatoric and epistemic components, providing a more robust and interpretable model. Comparative evaluations against conventional Long Short-Term Memory (LSTM)-based autoencoder and forecasting methods demonstrate that the proposed framework achieves superior fault-detection performance across multiple fault types, including misalignment, bearing failure, and unbalance. Moreover, uncertainty analyses confirm that fault severity correlates with increasing levels of both aleatoric and epistemic uncertainty, reflecting heightened noise and reduced model confidence under more severe conditions. By enhancing GAT fundamentals with a domain-tailored dual-graph strategy, specialized Bayesian inference, and digital-twin data generation, this research delivers a comprehensive and interpretable solution for compressor fault diagnosis, paving the way for more reliable and risk-aware predictive maintenance in complex rotating machinery. Full article

The intersection of green finance and artificial intelligence (AI) represents a rapidly emerging and high-impact research domain with the potential to reshape sustainable economic systems. This study presents a comprehensive bibliometric and network analysis aimed at mapping the scientific landscape, identifying research hotspots, and highlighting methodological trends at this nexus. A dataset of 268 peer-reviewed publications (2014–June 2025) was retrieved from the Web of Science Core Collection, filtered by the Business Economics category. Analytical techniques employed include Bibliometrix in R, VOSviewer, and science mapping tools such as thematic mapping, trend topic analysis, co-citation networks, and co-occurrence clustering. Results indicate an annual growth rate of 53.31%, with China leading in both productivity and impact, followed by Vietnam and the United Kingdom. The most prolific affiliations and authors, primarily based in China, underscore a concentrated regional research output. The most relevant journals include Energy Economics and Finance Research Letters. Network visualizations identified 17 clusters, with focused analysis on the top three: (1) Emission, Health, and Environmental Risk, (2) Institutional and Technological Infrastructure, and (3) Green Innovation and Sustainable Urban Development. The methodological landscape is equally diverse, with top techniques including blockchain technology, large language models, convolutional neural networks, sentiment analysis, and structural equation modeling, demonstrating a blend of traditional econometrics and advanced AI. This study not only uncovers intellectual structures and thematic evolution but also identifies underdeveloped areas and proposes future research directions. These include dynamic topic modeling, regional case studies, and ethical frameworks for AI in sustainable finance. The findings provide a strategic foundation for advancing interdisciplinary collaboration and policy innovation in green AI–finance ecosystems. Full article

Background: The use of lower limb prosthesis can impact all aspects of daily life, activities and participation. Various studies have compared the microprocessor-controlled knee (MPK) to the non-microprocessor-controlled knee (NMPK) using a variety of different outcome measures, but results are inconsistent and raise the question of which type of knee is most effective. Therefore, we aimed to assess the effect of MPKs compared to NMPKs across all classified ICF domains in adult prosthesis users. Methods: Participants performed baseline measurements with the NMPK (T0). One week later, they started a four-to-six-week trial period with the MPK. Afterward, measurements were repeated with the MPK (T1). Functional tests (6MWT, TUG-test and activity monitor) and questionnaires (ABC, SQUASH, USER-P and PEQ) were used. For statistical analyses, paired t-tests, Wilcoxon signed-rank tests and Chi² test were applied. The Benjamini–Hochberg procedure was applied to correct for multiple testing. Results: Twenty-five participants were included. Using an MPK compared to an NMPK significantly resulted in improvements in balance and walking confidence, safety, walking distance and self-reported walking ability, as well as a decrease in number of stumbles and falls. Additionally, participants using an MPK were significantly more satisfied with their participation, experienced fewer restrictions, reported greater satisfaction with the appearance and utility of the MPK, experienced less social burden and reported better well-being, compared to using an NMPK. Conclusions: Using an MPK instead of an NMPK can lead to significant improvements in all classified ICF domains, such as improved walking ability, confidence and satisfaction and reduced fall risk. Full article

While IPV is often studied as a predictor of housing insecurity, few U.S. studies explore how different forms of housing instability may contribute to intimate partner violence (IPV) risk. Using a mixed-methods approach and a cross-sectional design, this study examined the association between four housing instability domains and IPV among a sample of tenants that had either experienced eviction or were at high risk for eviction. Tenants in Harris and Travis counties (Texas, USA) completed an online survey (n = 1085; March–July 2024). Housing instability was assessed across four domains: homelessness, lease violations, utility hardship, and poor housing quality. IPV was measured using the Hurt, Insult, Threaten, Scream Screener. Covariate-adjusted logistic regression models suggest indicators within the four housing instability domains were associated with IPV risk. Within the homelessness domain, experiences with lifetime homelessness (AOR = 1.92, 95%CI 1.61–2.28), in the past 12 months living in unconventional spaces (AOR = 2.10, 95%CI 1.92–2.29), and moving in with others (AOR = 1.20, 95%CI 1.06–1.36) were associated with IPV. Within the lease violations domain, missed rent payments (AOR = 1.69, 95%CI 1.68–1.71) and non-payment lease violations (AOR = 2.50, 95%CI 2.29–2.73) in the past 12 months were associated with IPV. Utility shutoffs (AOR = 1.62, 95%CI 1.37–1.91) and unsafe housing (AOR = 1.65, 95%CI 1.31–2.09) in the past 12 months were associated with IPV. Homelessness, housing-related economic hardships and substandard living conditions predict an elevated risk of IPV. Full article

This study developed a model to predict ships with high deficiency risk under Port State Control (PSC) through machine learning techniques, particularly the Random Forest algorithm. The study utilized actual ship inspection data from the Port of Singapore, comprehensively considering various operational and safety indicators of ships, including but not limited to flag state, ship age, past deficiencies, and detention history. By analyzing these factors in depth, this research enhances the efficiency and accuracy of PSC inspections, provides decision support for port authorities, and offers strategic guidance for shipping companies to comply with international safety standards. During the research process, I first conducted detailed data preprocessing, including data cleaning and feature selection, to ensure the effectiveness of model training. Using the Random Forest algorithm, I identified key factors influencing the detention risk of ships and established a risk prediction model accordingly. The model validation results indicated that factors such as ship age, tonnage, company performance, and flag state significantly affect whether a ship exhibits a high deficiency rate. In addition, this study explored the potential and limitations of applying the Random Forest model in predicting high deficiency risk under PSC, and proposed future research directions, including further model optimization and the development of real-time prediction systems. By achieving these goals, I hope to provide valuable experience for other global shipping hubs, promote higher international maritime safety standards, and contribute to the sustainable development of the global shipping industry. This research not only highlights the importance of machine learning in the maritime domain but also demonstrates the potential of data-driven decision-making in improving ship safety management and port inspection efficiency. It is hoped that this study will inspire more maritime practitioners and researchers to explore advanced data analytics techniques to address the increasingly complex challenges of global shipping. Full article

Background: Lung cancer ranks as the leading cause of cancer-related mortality worldwide. The complexity of tumor delineation, crucial for radiation therapy, requires expertise often unavailable in resource-limited settings. Artificial Intelligence (AI), particularly with advancements in deep learning (DL) and natural language processing (NLP), offers potential solutions yet is challenged by high false positive rates. Purpose: The Oncology Contouring Copilot (OCC) system is developed to leverage oncologist expertise for precise tumor contouring using textual descriptions, aiming to increase the efficiency of oncological workflows by combining the strengths of AI with human oversight. Methods: Our OCC system initially identifies nodule candidates from CT scans. Employing Language Vision Models (LVMs) like GPT-4V, OCC then effectively reduces false positives with clinical descriptive texts, merging textual and visual data to automate tumor delineation, designed to elevate the quality of oncology care by incorporating knowledge from experienced domain experts. Results: The deployment of the OCC system resulted in a 35.0% reduction in the false discovery rate, a 72.4% decrease in false positives per scan, and an F1-score of 0.652 across our dataset for unbiased evaluation. Conclusions: OCC represents a significant advance in oncology care, particularly through the use of the latest LVMs, improving contouring results by (1) streamlining oncology treatment workflows by optimizing tumor delineation and reducing manual processes; (2) offering a scalable and intuitive framework to reduce false positives in radiotherapy planning using LVMs; (3) introducing novel medical language vision prompt techniques to minimize LVM hallucinations with ablation study; and (4) conducting a comparative analysis of LVMs, highlighting their potential in addressing medical language vision challenges. Full article

Forestry carbon sinks play a crucial role in mitigating climate change and protecting ecosystems, significantly contributing to the development of carbon trading systems. Remote sensing technology has become increasingly important for monitoring carbon sinks, as it allows for precise measurement of carbon storage and ecological changes, which are vital for forecasting carbon prices. Carbon prices fluctuate due to the interaction of various factors, exhibiting non-stationary characteristics and inherent uncertainties, making accurate predictions particularly challenging. To address these complexities, this study proposes a method for predicting carbon trading prices influenced by multiple factors. We introduce a Decomposition (DECOMP) module that separates carbon price data and its influencing factors into trend and cyclical components. To manage non-stationarity, we propose the KAN with Multi-Domain Diffusion (KAN-MD) module, which efficiently extracts relevant features. Furthermore, a Wave-MH attention module, based on wavelet transformation, is introduced to minimize interference from uncertainties, thereby enhancing the robustness of the model. Empirical research using data from the Hubei carbon trading market demonstrates that our model achieves superior predictive accuracy and resilience to fluctuations compared to other benchmark methods, with an MSE of 0.204% and an MAE of 0.0277. These results provide reliable support for pricing carbon financial derivatives and managing associated risks. Full article

This paper introduces a theoretical large-scale radio channel model for the downlink in cellular systems, aimed at estimating variations in received signal power at the user terminal as a function of device mobility. This enables applications such as direction-of-arrival (DoA) estimation, estimating power at subsequent points based on received power, and detection of coverage anomalies. The model is validated using real-world measurements from urban and suburban environments, achieving a maximum estimation error of 7.6%. In contrast to conventional models like Okumura–Hata, COST-231, Third Generation Partnership Project (3GPP) stochastic models, or ray-tracing techniques, which estimate average power under static conditions, the proposed model captures power fluctuations induced by terminal movement, a factor often neglected. Although advanced techniques such as wave-domain processing with intelligent metasurfaces can also estimate DoA, this model provides a simpler, geometry-driven approach based on empirical traces. While it does not incorporate infrastructure-specific characteristics or inter-cell interference, it remains a practical solution for scenarios with limited information or computational resources. Full article

The growth and sustainability of today’s global economy heavily relies on smooth maritime operations. The increasing security concerns to marine environments pose complex security challenges, such as smuggling, illegal fishing, human trafficking, and environmental threats, for traditional surveillance methods due to their limitations. Artificial intelligence (AI), particularly deep learning, has offered strong capabilities for automating object detection, anomaly identification, and situational awareness in maritime environments. In this paper, we have reviewed the state-of-the-art deep learning models mainly proposed in recent literature (2020–2025), including convolutional neural networks, recurrent neural networks, Transformers, and multimodal fusion architectures. We have highlighted their success in processing diverse data sources such as satellite imagery, AIS, SAR, radar, and sensor inputs from UxVs. Additionally, multimodal data fusion techniques enhance robustness by integrating complementary data, yielding more detection accuracy. There still exist challenges in detecting small or occluded objects, handling cluttered scenes, and interpreting unusual vessel behaviours, especially under adverse sea conditions. Additionally, explainability and real-time deployment of AI models in operational settings are open research areas. Overall, the review of existing maritime literature suggests that deep learning is rapidly transforming maritime domain awareness and response, with significant potential to improve global maritime security and operational efficiency. We have also provided key datasets for deep learning models in the maritime security domain. Full article

In this study, we focus on solving the nonlinear time-fractional Hirota–Satsuma coupled Korteweg–de Vries (KdV) and modified Korteweg–de Vries (MKdV) equations, using the Yang transform iterative method (YTIM). This method combines the Yang transform with a new iterative scheme to construct reliable and efficient solutions. Readers can understand the procedures clearly, since the implementation of Yang transform directly transforms fractional derivative sections into algebraic terms in the given problems. The new iterative scheme is applied to generate series solutions for the provided problems. The fractional derivatives are considered in the Caputo sense. To validate the proposed approach, two numerical examples are analysed and compared with exact solutions, as well as with the results obtained from the fractional reduced differential transform method (FRDTM) and the q-homotopy analysis transform method (q-HATM). The comparisons, presented through both tables and graphical illustrations, confirm the enhanced accuracy and reliability of the proposed method. Moreover, the effect of varying the fractional order is explored, demonstrating convergence of the solution as the order approaches an integer value. Importantly, the time-fractional Hirota–Satsuma coupled KdV and modified Korteweg–de Vries (MKdV) equations investigated in this work are not only of theoretical and computational interest but also possess significant implications for achieving global sustainability goals. Specifically, these equations contribute to the Sustainable Development Goal (SDG) “Life Below Water” by offering advanced modelling capabilities for understanding wave propagation and ocean dynamics, thus supporting marine ecosystem research and management. It is also relevant to SDG “Climate Action” as it aids in the simulation of environmental phenomena crucial to climate change analysis and mitigation. Additionally, the development and application of innovative mathematical modelling techniques align with “Industry, Innovation, and Infrastructure” promoting advanced computational tools for use in ocean engineering, environmental monitoring, and other infrastructure-related domains. Therefore, the proposed method not only advances mathematical and numerical analysis but also fosters interdisciplinary contributions toward sustainable development. Full article

The design and adjustable properties of nanoporous materials are important for current and future technological applications, research, and development. In addition, nanoporous ferroelectric materials have the potential to achieve competitive ferroelectric, dielectric, and piezoelectric characteristics. In this work, using the phase-field model, we found that the shape of pores in barium titanite ceramics governs the formation of the ferroelectric domain structure and the switching hysteresis loop. A remanent polarization-coercive field (P_r-E_c) diagram is introduced to denote the shape of the hysteresis loops. We performed a fundamental study in understanding how the domain structures affect the properties of domain-engineered porous ferroelectrics. Simulation results show that the hysteresis loop of porous ferroelectrics can be designed by controlling the shape/orientation of the ellipse-shaped pores. Numerical simulations also verify that the dielectric/piezoelectric properties can be improved with artificially designed porous structures. These phase-field results may be useful in the development of highly performing lead-free ferroelectric/piezoelectric materials. Full article