Search Results (1,942)

Reliability assessment of power systems is essential for ensuring the secure and stable operation of power grids, and identifying weak links constitutes a critical step in enhancing system reliability. Traditional deterministic methods are limited in their ability to capture the complex, nonlinear relationships between component failures and overall system risk. To overcome this limitation, this paper proposes an explainable machine learning-based approach for identifying weak components in power systems. Specifically, a set of contingency scenarios is constructed through enumeration, and a random forest regression model is trained to map transmission line outage events to the amount of system load curtailment. The trained model is then interpreted using SHapley Additive exPlanations (SHAP) values. By aggregating these values, the global reliability contribution of each component is quantified. The proposed method is validated on the IEEE 57-bus system, and the results demonstrate its effectiveness and feasibility. This research offers a data-driven framework for translating system-level reliability metrics into device-level quantitative attributions, thereby enabling interpretable identification of weak links. Full article

Accurate recognition of basic facial emotions is well documented, yet the mechanisms of misclassification and their relation to gaze allocation remain under-reported. The present study utilized a within-subjects eye-tracking design to examine both accurate and inaccurate recognition of five basic emotions (anger, disgust, fear, happiness, and sadness) in healthy young adults. Fifty participants (twenty-four women) completed a forced-choice categorization task with 10 stimuli (female/male poser × emotion). A remote eye tracker (60 Hz) recorded fixations mapped to eyes, nose, and mouth areas of interest (AOIs). The analyses combined accuracy and decision-time statistics with heatmap comparisons of misclassified versus accurate trials within the same image. Overall accuracy was 87.8% (439/500). Misclassification patterns depended on the target emotion, but not on participant gender. Fear male was most often misclassified (typically as disgust), and sadness female was frequently labeled as fear or disgust; disgust was the most incorrectly attributed response. For accurate trials, decision time showed main effects of emotion (p < 0.001) and participant gender (p = 0.033): happiness was categorized fastest and anger slowest, and women responded faster overall, with particularly fast response times for sadness. The AOI results revealed strong main effects and an AOI × emotion interaction (p < 0.001): eyes received the most fixations, but fear drew relatively more mouth sampling and sadness more nose sampling. Crucially, heatmaps showed an upper-face bias (eye AOI) in inaccurate trials, whereas accurate trials retained eye sampling and added nose and mouth AOI coverage, which aligned with diagnostic cues. These findings indicate that the scanpath strategy, in addition to information availability, underpins success and failure in basic-emotion recognition, with implications for theory, targeted training, and affective technologies. Full article

Background: Alzheimer’s disease (AD) is a progressive neurodegenerative condition marked by cognitive decline and memory loss. Despite advancements in AI-driven neuroimaging analysis for AD detection, clinical deployment remains limited due to challenges in model interpretability and usability. Explainable AI (XAI) frameworks such as XRAI offer potential to bridge this gap by providing clinically meaningful visualizations of model decision-making. Methods: This study developed a comprehensive, clinically deployable AI system for AD severity classification using 2D brain MRI data. Three deep learning architectures MobileNet-V3 Large, EfficientNet-B4, and ResNet-50 were trained on an augmented Kaggle dataset (33,984 images across four AD severity classes). The models were evaluated on both augmented and original datasets, with integrated XRAI explainability providing region-based attribution maps. A web-based clinical interface was built using Gradio to deliver real-time predictions and visual explanations. Results: MobileNet-V3 achieved the highest accuracy (99.18% on the augmented test set; 99.47% on the original dataset), while using the fewest parameters (4.2 M), confirming its efficiency and suitability for clinical use. XRAI visualizations aligned with known neuroanatomical patterns of AD progression, enhancing clinical interpretability. The web interface delivered sub-20 s inference with high classification confidence across all AD severity levels, successfully supporting real-world diagnostic workflows. Conclusions: This research presents the first systematic integration of XRAI into AD severity classification using MRI and deep learning. The MobileNet-V3-based system offers high accuracy, computational efficiency, and interpretability through a user-friendly clinical interface. These contributions demonstrate a practical pathway toward real-world adoption of explainable AI for early and accurate Alzheimer’s disease detection. Full article

Commercial computer-aided design (CAD) software is often expensive. This paper examines the use of product manufacturing information (PMI) web visualization to address the challenges faced by production site personnel and external partners collaborating on product development. These individuals need to be able to view or query PMI in model-based definition models without having to install professional CAD software. A detailed analysis of the relationships between PMI entity attributes in standard for the exchange of product model data (STEP) AP242 files was conducted. An algorithm for the automatic parsing and mapping of PMI semantics to a web browser is presented. Using linear sizes as an example, this paper introduces a prototype system with the following features: PMI web visualization; automatic linkage of PMI to associated geometry; browser-native rendering without the need for dedicated applications; and integration of graphical presentation and semantic representation. The effectiveness and feasibility of the prototype system are validated through case studies. However, the system has limitations when handling large assemblies with compound tolerances, curved dimension placements, and overlapping annotations, which presents areas for future research. Full article

Accurate assessment of coalbed methane (CBM) content is essential for characterizing subsurface reservoir distribution, guiding well placement, and estimating reserves. Current methods for determining coal seam gas content mainly rely on direct laboratory measurements of core samples or indirect interpretations derived from well log data. However, conventional coring is costly, while log-based approaches often depend on linear empirical formulas and are restricted to near-wellbore regions. In practice, the relationships between elastic properties and gas content are highly complex and nonlinear, leading conventional linear models to produce substantial prediction errors and inadequate performance. This study introduces a novel method for predicting gas content in deep coal seams using a Conditional Generative Adversarial Network (CGAN). First, elastic parameters are obtained through pre-stack inversion. Next, sensitivity analysis and attribute optimization are applied to identify elastic attributes that are most sensitive to gas content. A CGAN is then employed to learn the nonlinear mapping between multiple fluid-sensitive seismic attributes and gas content distribution. By integrating multiple constraints to refine the discriminator and guide generator training, the model achieves accurate gas content prediction directly from seismic data. Applied to a real dataset from a CBM block in the Ordos Basin, China, the proposed CGAN-based method produces predictions that align closely with measured gas content trends at well locations. Validation at blind wells shows an average prediction error of 1.6 m³/t, with 83% of samples exhibiting errors less than 3 m³/t. This research presents an effective and innovative deep learning approach for predicting coalbed methane content. Full article

Chile boasts a rich variety of native endemic melliferous flora, recognized internationally for the excellent taste and biological properties of its honeys. While honey production occurs across various regions, the southern zone, particularly near native rainforests, yields highly valued honeys that often lack comprehensive analytical characterization. This study was focused on seven apiaries near Lake Ranco in the Los Rios Region, collecting two honey samples from each location over two consecutive harvesting seasons, totaling 20 samples. Key parameters analyzed included botanical origin, total carbohydrates, glucose/fructose ratio, total phenolic compounds, antioxidant/antiradical activity and 5-hydroxymethylfurfural (HMF) content. The results indicated a significant influence of three native species, Eucryphia cordifolia, Caldcluvia paniculata, and Weinmannia trichosperma, on the antioxidant activity of the honeys. The physicochemical parameters measured, along with the concentration and activity of the compounds responsible for this activity, establish a very characteristic pattern for the monofloral honeys of these three species. This information could serve as a foundation for constructing a map to help differentiate Chilean honeys based on their natural biological attributes helpful for consumer health, generating distinctive profiles that would contribute to accurately guaranteeing their geographical origin and, consequently, increase their specific value. Full article

Large Language Models (LLMs) exhibit remarkable generative capabilities but remain vulnerable to hallucinations—outputs that are fluent yet inaccurate, ungrounded, or inconsistent with source material. To address the lack of methodologically grounded surveys, this paper introduces a novel method-oriented taxonomy of hallucination mitigation strategies in text-based LLMs. The taxonomy organizes over 300 studies into six principled categories: Training and Learning Approaches, Architectural Modifications, Input/Prompt Optimization, Post-Generation Quality Control, Interpretability and Diagnostic Methods, and Agent-Based Orchestration. Beyond mapping the field, we identify persistent challenges such as the absence of standardized evaluation benchmarks, attribution difficulties in multi-method systems, and the fragility of retrieval-based methods when sources are noisy or outdated. We also highlight emerging directions, including knowledge-grounded fine-tuning and hybrid retrieval–generation pipelines integrated with self-reflective reasoning agents. This taxonomy provides a methodological framework for advancing reliable, context-sensitive LLM deployment in high-stakes domains such as healthcare, law, and defense. Full article

To address the fragmentation and weak correlation of knowledge in the design method domain, this paper proposes a framework for constructing a knowledge graph driven by multi-source heterogeneous data. The process involves collecting multi-source heterogeneous data and subsequently utilizing text mining and natural language processing techniques to extract design themes and method elements. A “theme–stage–attribute” three-dimensional mapping model is established to achieve semantic coupling of knowledge. The BERT-BiLSTM-CRF (Bidirectional Encoder Representations from Transformers-Bidirectional Long Short-Term Memory-Conditional Random Field) model is employed for entity recognition and relation extraction, while the Sentence-BERT (Sentence Bidirectional Encoder Representations from Transformers) model is used to perform multi-source knowledge fusion. The Neo4j graph database facilitates knowledge storage, visualization, and querying, forming the basis for developing a prototype of a design method recommendation system. The framework’s effectiveness was validated through experiments on extraction performance and knowledge graph quality. The results demonstrate that the framework achieves an F1 score of 91.2% for knowledge extraction, and an 8.44% improvement over the baseline. The resulting graph’s node and relation coverage reached 94.1% and 91.2%, respectively. In complex semantic query tasks, the framework shows a significant advantage over traditional classification systems, achieving a maximum F1 score of 0.97. It can effectively integrate dispersed knowledge in the field of design methods and support method matching throughout the entire design process. This research is of significant value for advancing knowledge management and application in innovative product design. Full article

Driving wheel slippage in agricultural tractors is influenced by soil moisture, density, and penetration resistance. These surface variations reflect post-tillage composition, enabling dynamic mapping via Remotely Piloted Aircraft (RPAs). This study evaluated ballast recommendations based on soil surface data and slippage percentages, correlating added wheel weights at different speeds for a tractor-reversible plow system. Six 94.5 m² quadrants were analyzed for slippage monitored by RPA (Mavic3M-RTK) pre- and post-agricultural operation overflights and soil sampling (moisture, density, penetration resistance). A 2 × 2 factorial scheme (F-test) assessed soil-surface attribute correlations and slippage under varying ballasts (52.5–57.5 kg/hp) and speeds. Results showed slippage ranged from 4.06% (52.5 kg/hp, fourth reduced gear) to 11.32% (57.5 kg/hp, same gear), with liquid ballast and gear selection significantly impacting performance in friable clayey soil. Digital Elevation Model (DEM) and spectral indices derived from RPA imagery, including Normalized Difference Red Edge (NDRE), Normalized Difference Water Index (NDWI), Bare Soil Index (BSI), Green–Red Vegetation Index (GRVI), Visible Atmospherically Resistant Index (VARI), and Slope, proved effective. The approach reduced tractor slippage from 11.32% (heavy ballast, 4th gear) to 4.06% (moderate ballast, 4th gear), showing clear improvement in traction performance. The integration of indices and slope metrics supported ballast adjustment strategies, particularly for secondary plowing operations, contributing to improved traction performance and overall operational efficiency. Full article

Background: Lumbar disk herniation is a common cause of low back pain and radiculopathy, significantly impacting patients’ life quality and functional capacity. Magnetic Resonance Imaging (MRI) remains the gold standard for its assessment due to its superior soft tissue contrast and multiplanar imaging capabilities. However, recent advances in spectral computed tomography (CT), particularly dual-energy CT (DECT), have introduced new diagnostic opportunities, offering improved soft tissue characterization. Objective: To evaluate the diagnostic performance of DECT in detecting and grading lumbar disk herniations using dedicated color-coded fat maps. Materials and Methods: A total of 205 intervertebral levels from 41 consecutive patients with lumbar symptoms were prospectively analyzed. All patients underwent both DECT and MRI within 3 days. Three radiologists with varying years of experience independently assessed DECT images using color-coded reconstructions. A five-point grading score was attributed to each lumbar level: 1 = normal disk, 2 = bulging/protrusion, 3 = focal herniation, 4 = extruded herniation, and 5 = migrated fragment. The statistical analysis included Pearson’s correlation for score consistency, Cohen’s Kappa for interobserver agreement, generalized estimating equations for a cluster-robust analysis, and an ROC curve analysis. The DECT diagnostic accuracy was assessed in a dichotomized model (grades 1–2 = no herniation; 3–5 = herniation), using MRI as reference. Results: A strong correlation was observed between DECT and MRI scores across all readers (mean Pearson’s r = 0.826, p < 0.001). The average exact agreement between DECT and MRI was 79.4%, with the highest concordance at L1–L2 (86.7%) and L5–S1 (80.4%). The interobserver agreement was substantial (mean Cohen’s κ = 0.765), with a near-perfect agreement between the two most experienced readers (κ = 0.822). The intraclass correlation coefficient was 0.906 (95% CI: 0.893–0.918). The ROC analysis showed excellent performance (AUC range: 0.953–0.986). In the dichotomous model, DECT demonstrated a markedly higher sensitivity than conventional CT (95.1% vs. 57.2%), with a comparable specificity (DECT: 99.0%; CT: 96.5%) and improved overall accuracy (98.4% vs. 90.0%). Subgroup analyses by age and disk location revealed no statistically significant differences. Conclusions: The use of DECT dedicated color-coded fat map reconstructions showed high diagnostic performance in the assessment of lumbar disk herniations compared to MRI. These findings support the development of dedicated post-processing tools, facilitating the broader clinical adoption of spectral CT, especially in cases where MRI is contraindicated or less accessible. Full article

The Attribute-Based Access Control (ABAC) model provides access control decisions based on subject, object (resource), and contextual attributes. However, the use of sensitive attributes in access control decisions poses many security and privacy challenges, particularly in cloud environment where third parties are involved. To address this shortcoming, we present a novel privacy-preserving Dummy-ABAC model that obfuscates real attributes with dummy attributes before transmission to the cloud server. In the proposed model, only dummy attributes are stored in the cloud database, whereas real attributes and mapping tokens are stored in a local machine database. Only dummy attributes are used for the access request evaluation in the cloud, and real data are retrieved in the post-decision mechanism using secure tokens. The security of the proposed model was assessed using a simulated threat scenario, including attribute inference, policy injection, and reverse mapping attacks. Experimental evaluation using machine learning classifiers (“DecisionTree” DT, “RandomForest” RF), demonstrated that inference accuracy dropped from ~0.65 on real attributes to ~0.25 on dummy attributes confirming improved resistance to inference attacks. Furthermore, the model rejects malformed and unauthorized policies. Performance analysis of dummy generation, token generation, encoding, and nearest-neighbor search, demonstrated minimal latency in both local and cloud environments. Overall, the proposed model ensures an efficient, secure, and privacy-preserving access control in cloud environments. Full article

Rice is one of the most important staple crops in China, and the rapid and accurate extraction of rice planting areas plays a crucial role in the agricultural management and food security assessment. However, the existing rice field identification methods faced the significant challenges in mountainous regions due to the severe cloud contamination, insufficient utilization of multi-dimensional features, and limited classification accuracy. This study presented a novel rice field identification method based on the Graph Convolutional Networks (GCN) that effectively integrated multi-source remote sensing data tailored for the complex mountainous terrain. A coarse-to-fine cloud removal strategy was developed by fusing the synthetic aperture radar (SAR) imagery with temporally adjacent optical remote sensing imagery, achieving high cloud removal accuracy, thereby providing reliable and clear optical data for the subsequent rice mapping. A comprehensive multi-feature library comprising spectral, texture, polarization, and terrain attributes was constructed and optimized via a stepwise selection process. Furthermore, the 19 key features were established to enhance the classification performance. The proposed method achieved an overall accuracy of 98.3% for the rice field identification in Huoshan County of the Dabie Mountains, and a 96.8% consistency compared to statistical yearbook data. The ablation experiments demonstrated that incorporating terrain features substantially improved the rice field identification accuracy under the complex topographic conditions. The comparative evaluations against support vector machine (SVM), random forest (RF), and U-Net models confirmed the superiority of the proposed method in terms of accuracy, local performance, terrain adaptability, training sample requirement, and computational cost, and demonstrated its effectiveness and applicability for the high-precision rice field distribution mapping in mountainous environments. Full article

When a city decides to undertake a certain urban project, is it modifying just the physical environment or the social fabric that dwells within? This work investigates the relationship between the geometric configuration of urban space (geometry–city) and the topology of the networks of encounters of its inhabitants (network–city) that form through daily interactions. The research departs from the hypothesis that changes in geometry–city would not significantly alter the topology of the network–city, testing this proposition conceptually through abstract computational simulations developed specifically for this study. In this simulator, abstract maps with buildings distributed over different primary geometries are generated and have activities (use: home or work) and a population assigned. Encounters of the “inhabitants” are registered while daily commute routines, enough to achieve differentiation and stability, are run. The initial results revealed that the geometry description was not enough, and definitions regarding activity attribution were also necessary. Thus, we could not confirm nor reject the original hypothesis exactly, but it had to be complemented, including the idea of an activity–city dimension. We found that despite the geometry–city per se not determining the structure of the network–city, the spatial (geometric) distribution of activities directly impacts the resulting topology. Urban geometry influences networks–city only insofar as it conforms to activity–city, defining areas for activities or restricting routing between them. But it is the geometry of localization of the activities that has a direct impact on the topology of the network–city. This conceptual discovery can have significant implications for urban planning if corroborated in real-world situations. It could suggest that land use policies may be more effective for intervening in network-based characteristics, like social cohesion and resilience, than purely morphological interventions. Full article

Background: Gemcitabine (GEM) remains a cornerstone in the treatment of pancreatic cancer. Upon exposure to GEM, pancreatic cancer cells (PCCs) tend to adapt quickly to outcompete drug-induced cytotoxicity, thereby contributing to treatment failure. Thus, understanding GEM-induced molecular changes in PCCs is important. Methods: Three primary PCC lines (PCC-1, PCC-2, PCC-7) and Mia PaCa-2 cultured for 40 passages (p) in the absence (control) or presence of GEM (GemR) were assessed for phenotypic changes. Proteome profiles for all PCCs at p10, p20, p25, p30, p35, and p40 were obtained using mass spectrometry (MS). Protein expression was determined using immunoblotting. Differentially abundant proteins (DAPs) were evaluated for enrichment of functional and biological attributes and protein–protein interactions. Results: GEM sensitivity and growth were both reduced in GemR versus paired controls for all four PCC lines. MS mapped > 7000 proteins in each PCC line, and the abundance of 70–83% of these was found to be significantly altered when comparing all sample groups. Proteomic changes in GemR versus paired controls differed remarkably among the PCCs and were affected by passaging and treatment duration. DAPs at p40 were mostly related to metabolic pathways, including nucleotide metabolism and diverse cell growth processes. Several closely related DAPs and multiple hub proteins in each PCC line were identified. Conclusions: Overall, this study revealed cell-line-specific, heterogeneous changes in proteome profiles of PCCs following their long-term exposure to GEM, and these were likely affected by treatment duration, dosage, and passaging. Full article

Land degradation in Iraq’s Mesopotamian plain threatens food security and rural livelihoods, yet the relative roles of climatic water deficits versus anthropogenic pressures remain poorly attributed in space. We test the hypothesis that multi-timescale climatic water deficits (SPEI-03/-06/-12) exert a stronger effect on vegetation degradation risk than anthropogenic pressures, conditional on hydrological connectivity and irrigation. Using Babil and Al-Qadisiyah (2000–2023) as a case, we implement a four-part pipeline: (i) Fractional Vegetation Cover with Mann–Kendall/Sen’s slope to quantify greening/browning trends; (ii) LandTrendr to extract disturbance timing and magnitude; (iii) annual LULC maps from a Random Forest classifier to resolve transitions; and (iv) an XGBoost classifier to map degradation risk and attribute climate vs. anthropogenic influence via drop-group permutation (ΔAUC), grouped SHAP shares, and leave-group-out ablation, all under spatial block cross-validation. Driver attribution shows mid-term and short-term drought (SPEI-06, SPEI-03) as the strongest predictors, and conditional permutation yields a larger average AUC loss for the climate block than for the anthropogenic block, while grouped SHAP shares are comparable between the two, and ablation suggests a neutral to weak anthropogenic edge. The XGBoost model attains AUC = 0.884 (test) and maps 9.7% of the area as high risk (>0.70), concentrated away from perennial water bodies. Over 2000–2023, LULC change indicates CA +515 km², HO +129 km², UL +70 km², BL −697 km², WB −16.7 km². Trend analysis shows recovery across 51.5% of the landscape (+29.6% dec⁻¹ median) and severe decline over 2.5% (−22.0% dec⁻¹). The integrated design couples trend mapping with driver attribution, clarifying how compounded climatic stress and intensive land use shape contemporary desertification risk and providing spatial priorities for restoration and adaptive water management. Full article