Search Results (57)

With the widespread application of multi-view data across various domains, multi-view unsupervised feature selection (MUFS) has achieved remarkable progress in both feature selection (FS) and missing-view completion. However, existing MUFS methods typically rely on centralized servers, which not only fail to meet privacy requirements in distributed settings but also suffer from suboptimal FS quality and poor convergence. To overcome these challenges, we propose a novel federated incomplete MUFS method (Fed-IMUFS), which integrates a fractional Sparsity-Guided Whale Optimization Algorithm (SGWOA) and Tensor Alternating Learning (TAL). Within this federated learning framework, each client performs local optimization in two stages: in the first stage, SGWOA introduces an $L_{2,1}$ proximal projection to enforce row-sparsity in the FS weight matrix, while fractional-order dynamics and fractal-inspired elite kernel injection mechanisms enhance global search ability, yielding a discriminative and stable weight matrix; in the second stage, based on the obtained weight matrix, an alternating optimization framework with tensor decomposition is employed to iteratively complete missing views while simultaneously optimizing low-dimensional representations to preserve cross-view consistency, with the objective function gradually minimized until convergence. During federated training, the server employs an aggregation and distribution strategy driven by normalized mutual information, where clients upload only their local weight matrices and quality indicators, and the server adaptively fuses them into a global FS matrix before distributing it back to clients. This process achieves consistent FS across clients while safeguarding data privacy. Comprehensive evaluations on CEC2022 and several incomplete multi-view datasets confirm that Fed-IMUFS outperforms state-of-the-art methods, delivering stronger global optimization capability, higher-quality feature selection, faster convergence, and more effective handling of missing views. Full article

Accurate traffic flow prediction is vital for intelligent transportation systems, yet strong spatiotemporal coupling and multi-scale dynamics make modelling difficult. Existing methods often rely on static adjacency and short input windows, limiting adaptation to time-varying spatial relations and long-term patterns. To address these issues, we propose the Pre-trained Trend-aware Dynamic Graph Convolutional Network (PT-TDGCN), a two-stage framework. In the pre-training stage, a Transformer-based masked autoencoder learns segment-level temporal representations from historical sequences. In the prediction stage, three designs are integrated: (1) dynamic graph learning parameterized by tensor decomposition; (2) convolutional trend-aware attention that adds 1D convolutions to capture local trends while preserving global context; and (3) spatial graph convolution combined with lightweight fusion projection for aligning pre-trained, spatial, and temporal representations. Extensive experiments on four real-world datasets demonstrated that PT-TDGCN consistently outperformed 14 baseline models, achieving superior predictive accuracy and robustness. Full article

Symmetry plays a critical role in preserving the structural balance and statistical integrity of datasets, particularly in privacy-preserving machine learning. Differential privacy introduces random noise to individual data points to ensure privacy while maintaining the overall symmetry of statistical distributions. However, excessive noise can reduce the utility of data, model accuracy, and computational efficiency. This study proposes a symmetry-preserving optimization framework for differentially private machine learning by integrating feature importance and t-SNE (t-distributed Stochastic Neighbor Embedding), UMAP (Uniform Manifold Approximation and Projection), and PCA (Principal Component Analysis), respectively. Feature importance derived from a random forest selects high-value features to improve data relevance. At the same time, t-SNE preserves geometric symmetry by retaining local and global structures more effectively than PCA or UMAP. Therefore, t-SNE is the best feature extraction method for dimensionality reduction in the proposed scheme. Experimental results demonstrate that the t-SNE method significantly enhances model performance under differential privacy, showing improved accuracy and reduced computational time compared to PCA and UMAP while preserving the underlying symmetry of the data distributions. Full article

This paper proposes a Distributed Adaptive Regularization Algorithm (DARN) for solving composite non-convex and non-smooth optimization problems in multi-agent systems. The algorithm employs a three-phase iterative framework to achieve efficient collaborative optimization: (1) a local regularized optimization step, which utilizes proximal mappings to enforce strong convexity of weakly convex objectives and ensure subproblem well-posedness; (2) a consensus update based on doubly stochastic matrices, guaranteeing asymptotic convergence of agent states to a global consensus point; and (3) an innovative adaptive regularization mechanism that dynamically adjusts regularization strength using local function value variations to balance stability and convergence speed. Theoretical analysis demonstrates that the algorithm maintains strict monotonic descent under non-convex and non-smooth conditions by constructing a mixed time-scale Lyapunov function, achieving a sublinear convergence rate. Notably, we prove that the projection-based update rule for regularization parameters preserves lower-bound constraints, while spectral decay properties of consensus errors and perturbations from local updates are globally governed by the Lyapunov function. Numerical experiments validate the algorithm’s superiority in sparse principal component analysis and robust matrix completion tasks, showing a 6.6% improvement in convergence speed and a 51.7% reduction in consensus error compared to fixed-regularization methods. This work provides theoretical guarantees and an efficient framework for distributed non-convex optimization in heterogeneous networks. Full article

This study introduces Edible Streets as a distinct and scalable model of community-led urban food growing, specifically investigating the drivers and barriers to the initiative. Unlike traditional urban food-growing initiatives, Edible Streets explores the integration of edible plants into street verges and footpaths with direct community involvement of the people who live/work in a street. This study contributes new knowledge by evaluating Edible Streets through the COM-B model of behavioural change, through policy and governance in addition to behaviour change, and by developing practical frameworks to facilitate its implementation. Focusing on Oxford, the research engaged residents through 17 in-person interviews and 18 online surveys, alongside a stakeholder workshop with 21 policymakers, community leaders, and NGO representatives. Findings revealed strong motivation for Edible Streets, driven by values of sustainability, community resilience, and improved well-being. However, capability barriers, including knowledge gaps in gardening, land-use policies, and food preservation, as well as opportunity constraints related to land access, water availability, and environmental challenges, hindered participation. To address these, a How-to Guide was developed, and a pilot Edible Street project was launched. Future steps include establishing a licensing application model to facilitate urban food growing and conducting a Post-Use Evaluation and Impact Study. Nationally, this model could support Right to Grow policies, while globally, it aligns with climate resilience and food security goals. Locally grown food enhances biodiversity, reduces carbon footprints, and strengthens social cohesion. By tackling key barriers and scaling solutions, this study provides actionable insights for policymakers and practitioners to create resilient, equitable urban food systems. Full article

In the real world, most data are unlabeled, which drives the development of semi-supervised learning (SSL). Among SSL methods, least squares regression (LSR) has attracted attention for its simplicity and efficiency. However, existing semi-supervised LSR approaches suffer from challenges such as the insufficient use of unlabeled data, low pseudo-label accuracy, and inefficient label propagation. To address these issues, this paper proposes dual label propagation-driven least squares regression with feature selection, named DLPLSR, which is a pseudo-label-free SSL framework. DLPLSR employs a fuzzy-graph-based clustering strategy to capture global relationships among all samples, and manifold regularization preserves local geometric consistency, so that it implements the dual label propagation mechanism for comprehensive utilization of unlabeled data. Meanwhile, a dual-feature selection mechanism is established by integrating orthogonal projection for maximizing feature information with an ℓ_2,1-norm regularization for eliminating redundancy, thereby jointly enhancing the discriminative power. Benefiting from these two designs, DLPLSR boosts learning performance without pseudo-labeling. Finally, the objective function admits an efficient closed-form solution solvable via an alternating optimization strategy. Extensive experiments on multiple benchmark datasets show the superiority of DLPLSR compared to state-of-the-art LSR-based SSL methods. Full article

Hydraulic concrete is a critical infrastructure material, with its surface condition playing a vital role in quality assessments for water conservancy and hydropower projects. However, images taken in complex hydraulic environments often suffer from degraded quality due to low lighting, shadows, and noise, making it difficult to distinguish defects from the background and thereby hindering accurate defect detection and damage evaluation. In this study, following systematic analyses of hydraulic concrete color space characteristics, we propose a Dual-Branch Luminance–Chrominance Attention Network (DBLCANet-HCIE) specifically designed for low-light hydraulic concrete image enhancement. Inspired by human visual perception, the network simultaneously improves global contrast and preserves fine-grained defect textures, which are essential for structural analysis. The proposed architecture consists of a Luminance Adjustment Branch (LAB) and a Chroma Restoration Branch (CRB). The LAB incorporates a Luminance-Aware Hybrid Attention Block (LAHAB) to capture both the global luminance distribution and local texture details, enabling adaptive illumination correction through comprehensive scene understanding. The CRB integrates a Channel Denoiser Block (CDB) for channel-specific noise suppression and a Frequency-Domain Detail Enhancement Block (FDDEB) to refine chrominance information and enhance subtle defect textures. A feature fusion block is designed to fuse and learn the features of the outputs from the two branches, resulting in images with enhanced luminance, reduced noise, and preserved surface anomalies. To validate the proposed approach, we construct a dedicated low-light hydraulic concrete image dataset (LLHCID). Extensive experiments conducted on both LOLv1 and LLHCID benchmarks demonstrate that the proposed method significantly enhances the visual interpretability of hydraulic concrete surfaces while effectively addressing low-light degradation challenges. Full article

Masonry industrial buildings, common in the 19th and 20th centuries, represent a significant architectural typology. These structures are crucial to the study of industrial archeology, which focuses on preserving and revitalizing historical industrial heritage. Often left neglected and deteriorating, they hold great potential for adaptive reuse, transforming into vibrant cultural, commercial, or residential spaces through well-planned restoration and consolidation efforts. This paper explores a case study of such industrial architecture: a decommissioned factory near Naples. The complex consists of multiple structures with vertical supports made of yellow tuff stone and roofs framed by wooden trusses. To improve the building’s seismic resilience, a comprehensive analysis was conducted, encompassing its historical, geometric, and structural characteristics. Using advanced computer software, the factory was modelled with a macro-element approach, allowing for a detailed assessment of its seismic vulnerability. This approach facilitated both a global analysis of the building’s overall behaviour and the identification of potential local collapse mechanisms. Non-linear analyses revealed a critical lack of seismic safety, particularly in the Y direction, with significant out-of-plane collapse risk due to weak connections among walls. Based on these findings, a restoration and consolidation plan was developed to enhance the structural integrity of the building and to ensure its long-term safety and functionality. This plan incorporated metal tie rods, masonry strengthening through injections, and roof reconstruction. The proposed interventions not only address immediate seismic risks but also contribute to the broader goal of preserving this industrial architectural heritage. This study introduces a novel multidisciplinary methodology—integrating seismic analysis, traditional retrofit techniques, and sustainable reuse—specifically tailored to the rarely addressed typology of masonry industrial structures. By transforming the factory into a functional urban space, the project presents a replicable model for preserving industrial heritage within contemporary cityscapes. Full article

Seismic waveform feature extraction is a critical task in seismic exploration, as it directly impacts reservoir prediction and geological interpretation. However, large-scale seismic data and nonlinear relationships between seismic signals and reservoir properties are challenging for traditional machine learning methods. To address these limitations, this paper proposes a novel framework combining Convolutional Neural Network (CNN) and Uniform Manifold Approximation and Projection (UMAP) for seismic waveform feature extraction and analysis. The UMAP-CNN framework leverages the strengths of manifold learning and deep learning, enabling multi-scale feature extraction and dimensionality reduction while preserving both local and global data structures. The evaluation experiments, which considered runtime, receiver operating characteristic (ROC) curves, embedding distribution maps, and other quantitative assessments, illustrated that the UMAP-CNN outperformed t-distributed stochastic neighbor embedding (t-SNE), locally linear embedding (LLE) and isometric feature mapping (Isomap). A case study in the Ordos Basin further demonstrated that UMAP-CNN offers a high degree of accuracy in predicting coal seam thickness. Furthermore, our framework exhibited superior computational efficiency and robustness in handling large-scale datasets. Full article

Creating grayscale images from a color reality has been an inherent human practice since ancient times, but it became a technological challenge with the advent of the first black-and-white televisions and digital image processing. Decolorization is a process that projects visual information from a three-dimensional feature space to a one-dimensional space, thus reducing the dimensionality of the image while minimizing the loss of information. To achieve this, various strategies have been developed, including the application of color channel weights and the analysis of local and global image contrast, but there is no universal solution. In this paper, we propose a bio-inspired approach that combines findings from neuroscience on the architecture of the visual system and color coding with evidence from studies in the psychology of art. The goal is to simplify the decolorization process and facilitate its control through color-related concepts that are easily understandable to humans. This new method organizes colors in a scale that links activity on the retina with a system of opponent and complementary channels, thus allowing the adjustment of the perception of warmth and coolness in the image. The results show an improvement in chromatic contrast, especially in the warmth and coolness categories, as well as an enhanced ability to preserve subtle contrasts, outperforming other approaches in the Ishihara test used in color blindness detection. In addition, the method offers a computational advantage by reducing the process through direct pixel-level operation. Full article

Generative adversarial networks (GANs) prioritize pixel-level attributes over capturing the entire image distribution, which is critical in image synthesis. To address this challenge, we propose a dual-stream contrastive latent projection generative adversarial network (DSCLPGAN) for the robust augmentation of MRI images. The dual-stream generator in our architecture incorporates two specialized processing pathways: one is dedicated to local feature variation modeling, while the other captures global structural transformations, ensuring a more comprehensive synthesis of medical images. We used a transformer-based encoder–decoder framework for contextual coherence and the contrastive learning projection (CLP) module integrates contrastive loss into the latent space for generating diverse image samples. The generated images undergo adversarial refinement using an ensemble of specialized discriminators, where discriminator 1 (D1) ensures classification consistency with real MRI images, discriminator 2 (D2) produces a probability map of localized variations, and discriminator 3 (D3) preserves structural consistency. For validation, we utilized a publicly available MRI dataset which contains 3064 T1-weighted contrast-enhanced images with three types of brain tumors: meningioma (708 slices), glioma (1426 slices), and pituitary tumor (930 slices). The experimental results demonstrate state-of-the-art performance, achieving an SSIM of 0.99, classification accuracy of 99.4% for an augmentation diversity level of 5, and a PSNR of 34.6 dB. Our approach has the potential of generating high-fidelity augmentations for reliable AI-driven clinical decision support systems. Full article

Taiwan has been actively advancing offshore wind energy, with significant progress in deep-sea and large-scale turbine development. However, this growth poses challenges to coastal fishery communities, particularly regarding the protection of fishery rights and livelihoods. This study employs the Life Cycle Sustainability Assessment (LCSA) framework to evaluate the impact of offshore wind farm (OWF) on fishery rights in Taiwan. Through an extensive literature review, we identify key indicators influencing fishery rights within the OWF context. To ensure a comprehensive analysis, expert surveys from diverse fields provide additional insights into these impacts. By aligning our findings with international frameworks, the International Finance Corporation (IFC) Performance Standards (PS) and the Equator Principles (EP), this research underscores the significance of integrating both local concerns and global standards in OWF development. In the lifecycle of long-term, large-scale OWF projects, PS1 of the IFC PS is the most widely applicable standard, whereas P2, P4, P5 and P9 of the EP plays a central role in ensuring compliance and operational efficiency. This study uniquely integrates local fishery rights into global frameworks, bridging regional socio-economic concerns with international sustainability standards—a novel approach to balancing offshore wind development with community interests. Ultimately, this research emphasizes the importance of balancing renewable energy advancement with the preservation of fishery rights. Full article

Wind energy has emerged as a cornerstone in global efforts to transition to renewable energy, driven by its low environmental impact and significant generation potential. However, the inherent intermittency of wind, influenced by complex and dynamic atmospheric patterns, poses significant challenges for accurate wind speed prediction. Existing approaches, including statistical methods, machine learning, and deep learning, often struggle with limitations such as non-linearity, non-stationarity, computational demands, and the requirement for extensive, high-quality datasets. In response to these challenges, we propose a novel neighborhood preserving cross-dataset data augmentation framework for high-horizon wind speed prediction. The proposed method addresses data variability and dynamic behaviors through three key components: (i) the uniform manifold approximation and projection (UMAP) is employed as a non-linear dimensionality reduction technique to encode local relationships in wind speed time-series data while preserving neighborhood structures, (ii) a localized cross-dataset data augmentation (DA) approach is introduced using UMAP-reduced spaces to enhance data diversity and mitigate variability across datasets, and (iii) recurrent neural networks (RNNs) are trained on the augmented datasets to model temporal dependencies and non-linear patterns effectively. Our framework was evaluated using datasets from diverse geographical locations, including the Argonne Weather Observatory (USA), Chengdu Airport (China), and Beijing Capital International Airport (China). Comparative tests using regression-based measures on RNN, GRU, and LSTM architectures showed that the proposed method was better at improving the accuracy and generalizability of predictions, leading to an average reduction in prediction error. Consequently, our study highlights the potential of integrating advanced dimensionality reduction, data augmentation, and deep learning techniques to address critical challenges in renewable energy forecasting. Full article

In this paper, a fault detection method based on kernel global local preserving projection is presented to address the nonlinear characteristics of industrial systems. First, data are projected into a high-dimensional feature space through nonlinear mapping, enabling linear separability in this feature space. Subsequently, data features are extracted using the global local preserving projection method in the high-dimensional feature space. Finally, a monitoring model is established based on these features. Experiments on the Tennessee Eastman process and industrial boilers demonstrate that the proposed method balances global and local data structures, reduces nonlinear influences, and improves the fault detection rate. Full article

Traditional cereals, recognised for their adaptability, high nutritional value, and unique sensory characteristics, have largely been excluded from global food supply chains. Recent shifts in consumption patterns, particularly in urban areas, indicate a growing demand for high-quality bread, creating new opportunities for farmers interested in sustainable production techniques and traditional varieties. However, challenges such as seed availability, regulatory constraints, marketing strategies, and logistical barriers persist. This study, conducted within the framework of the CERTRA project—Development of Traditional Cereal Value Chains for Sustainable Food in Portugal—aims to enhance the traditional cereal value chain in Portugal by identifying key challenges and opportunities and proposing effective development strategies. The research employs a mixed-method approach, including documentary research, a SWOT analysis based on the scientific literature and stakeholder insights, and a case study methodology examining twelve successful European initiatives. The findings highlight strengths such as seed sovereignty, resilience under low-input farming, and market potential through certification and short food supply chains. However, weaknesses such as lower yields, mechanisation challenges, and seed access restrictions remain critical obstacles. Our analysis suggests that participatory breeding programs, farmer-led seed networks, and hybrid distribution models integrating direct sales, online platforms, and local partnerships can support the revitalisation of traditional cereals. Future research should focus on consumer preferences, branding strategies, and technological innovations that enhance processing efficiency while preserving the ecological and cultural value of traditional varieties. Full article