Search Results (162)

The increasing use of Arabic and Urdu on social media platforms, particularly Twitter, has created a growing need for robust Named Entity Recognition (NER) systems capable of handling noisy, informal, and code-mixed content. However, both languages remain significantly underrepresented in NER research, especially in social media contexts. To address this gap, this study makes four key contributions: (1) We introduced a manual entity consolidation step to enhance the consistency and accuracy of named entity annotations. In the original datasets, entities such as person names and organization names were often split into multiple tokens (e.g., first name and last name labeled separately). We manually refined the annotations to merge these segments into unified entities, ensuring improved coherence for both training and evaluation. (2) We selected two publicly available datasets from GitHub—one in Arabic and one in Urdu—and applied two novel strategies to tackle low-resource challenges: a joint multilingual approach and a translation-based approach. The joint approach involved merging both datasets to create a unified multilingual corpus, while the translation-based approach utilized automatic translation to generate cross-lingual datasets, enhancing linguistic diversity and model generalizability. (3) We presented a comprehensive and reproducible pseudocode-driven framework that integrates translation, manual refinement, dataset merging, preprocessing, and multilingual model fine-tuning. (4) We designed, implemented, and evaluated a customized XLM-RoBERTa model integrated with a novel attention mechanism, specifically optimized for the morphological and syntactic complexities of Arabic and Urdu. Based on the experiments, our proposed model (XLM-RoBERTa) achieves 0.98 accuracy across Arabic, Urdu, and multilingual datasets. While it shows a 7–8% improvement over traditional baselines (RF), it also achieves a 2.08% improvement over a deep learning (BiLSTM = 0.96), highlighting the effectiveness of our cross-lingual, resource-efficient approach for NER in low-resource, code-mixed social media text. Full article

Manual calibration of nuclear medicine scanners currently relies on handling phantoms containing radioactive sources, exposing personnel to high radiation doses and elevating cancer risk. We designed an automated detection framework for robotic inspection on the YOLOv8n foundation. It pairs a lightweight backbone with a shape-aware geometric attention module and an anchor-free head. Facing a small training set, we produced extra images with a GAN and then fine-tuned a pretrained network on these augmented data. Evaluations on a custom dataset consisting of PET/CT gantry and table images showed that the SAM-YOLOv8n model achieved a precision of 93.6% and a recall of 92.8%. These results demonstrate fast, accurate, real-time detection, offering a safer and more efficient alternative to manual calibration of nuclear medicine equipment. Full article

Currently, YOLO-based object detection is widely employed in intelligent inspection robots. However, under interference factors present in dimly lit substation environments, YOLO exhibits issues such as excessively low accuracy, missed detections, and false detections for critical targets. To address these problems, this paper proposes HSS-YOLO, a lightweight object detection model based on YOLOv11. Initially, HetConv is introduced. By combining convolutional kernels of different sizes, it reduces the required number of floating-point operations (FLOPs) and enhances computational efficiency. Subsequently, the integration of Inner-SIoU strengthens the recognition capability for small targets within dim environments. Finally, ShuffleAttention is incorporated to mitigate problems like missed or false detections of small targets under low-light conditions. The experimental results demonstrate that on a custom dataset, the model achieves a precision of 90.5% for critical targets (doors and two types of handles). This represents a 4.6% improvement over YOLOv11, while also reducing parameter count by 10.7% and computational load by 9%. Furthermore, evaluations on public datasets confirm that the proposed model surpasses YOLOv11 in assessment metrics. The improved model presented in this study not only achieves lightweight design but also yields more accurate detection results for doors and handles within dimly lit substation environments. Full article

This study examines the influence of different printing orientations and infill settings on the strength and flexibility of components produced using resin-based 3D printing, particularly with masked stereolithography (MSLA). Using a common photopolymer resin and a widely available desktop MSLA printer, we produced and tested a series of samples with varying tilt angles and internal structures. To understand their mechanical behavior, we applied a custom bending test combined with high-precision deformation tracking through the GOM ARAMIS digital image correlation system. The results obtained clearly show that both the angle of printing and the density of the internal infill structure play a significant role in how much strain the printed parts can handle before breaking. Notably, a 75° orientation provided the best deformation performance, and infill rates between 60% and 90% offered a good balance between strength and material efficiency. These findings highlight how adjusting print settings can lead to stronger parts while also saving time and resources—an important consideration for practical applications in engineering, design, and manufacturing. Full article

Remanufacturing design is a green design model that considers remanufacturability during the design process to improve the reuse of components. However, traditional remanufacturing design scheme decision making focuses on the remanufacturability indicator and does not fully consider the carbon emissions of the remanufacturing process, which will take away the energy-saving and emission reduction benefits of remanufacturing. In addition, remanufacturing design schemes rarely consider the human ergonomics of the product, which leads to uncomfortable handling of the product by the customer. To reduce the remanufacturing carbon emission and improve customer comfort, it is necessary to select a reasonable design scheme to satisfy the carbon emission reduction and ergonomics demand; therefore, this paper proposes an integrated multi-criteria decision-making method for remanufacturing design that considers the carbon emission and human ergonomics. Firstly, an evaluation system of remanufacturing design schemes is constructed to consider the remanufacturability, cost, carbon emission, and human ergonomics of the product, and the evaluation indicators are quantified by the normalization method and the Kansei engineering (KE) method; meanwhile, the hierarchical analysis method (AHP) and entropy weight method (EW) are used for the calculation of the subjective and objective weights. Then, a multi-attribute decision-making method based on the combination of an assignment approximation of ideal solution ranking (TOPSIS) and gray correlation analysis (GRA) is proposed to complete the design scheme selection. Finally, the feasibility of the scheme is verified by taking a household coffee machine as an example. This method has been implemented as an application using Visual Studio 2022 and Microsoft SQL Server 2022. The research results indicate that this decision-making method can quickly and accurately generate reasonable remanufacturing design schemes. Full article

Teaching of embedded systems, including communication protocols such as SMBus, is commonly faced with difficulties providing the students with interactive and personalized, practical learning experiences. To overcome these shortcomings, this report presents a new conceptual framework that exploits generative artificial intelligence (GenAI) via customized DV-GPT. Coupled with prepromises techniques, DV-GPT offers timely targeted support to students and engineers who are studying SMBus protocol design and verification. In contrast to traditional learning, this AI-based tool dynamically adjusts feedback based on the users’ activities, providing greater insight into challenging concepts, including timing synchronization, multi-master arbitration, and error handling. The framework also incorporates the industry de facto standard UVM practices, which helps narrow the gap between education and the professional world. We quantitatively compare with a baseline GPT-4 and show significant improvement in accuracy, specificity, and user satisfaction. The effectiveness and feasibility of the proposed GenAI-enhanced educational approach have been empirically validated through the use of structured student feedback, expert judgment, and statistical analysis. The contribution of this research is a scalable, flexible, interactive model for enhancing embedded systems education that also illustrates how GenAI technologies could find applicability within specialized educational environments. Full article

The accurate prediction of carbon dioxide (CO₂) adsorption in metal–organic frameworks (MOFs) is critical for accelerating the discovery of high-performance materials for post-combustion carbon capture. Experimental screening of MOFs is often costly and time-consuming, creating a strong incentive to develop reliable data-driven models. Despite extensive research, most studies rely on standalone models or generic ensemble strategies that fall short in handling the complex, nonlinear relationships inherent in adsorption data. In this study, a novel ensemble learning framework is developed by integrating five distinct regression algorithms: Random Forest, XGBoost, LightGBM, Support Vector Regression, and Multi-Layer Perceptron. These algorithms are combined into four custom ensemble strategies: equal-weighted voting, performance-weighted voting, stacking, and manual blending. A dataset comprising 1212 experimentally validated MOF entries with input descriptors including BET surface area, pore volume, pressure, temperature, and metal center is used to train and evaluate the models. The stacking ensemble yields the highest performance, with an R² of 0.9833, an RMSE of 1.0016, and an MAE of 0.6630 on the test set. Model reliability is further confirmed through residual diagnostics, prediction intervals, and permutation importance, revealing pressure and temperature to be the most influential features. Ablation analysis highlights the complementary role of all base models, particularly Random Forest and LightGBM, in boosting ensemble performance. This study demonstrates that custom ensemble learning strategies not only improve predictive accuracy but also enhance model interpretability, offering a scalable and cost-effective tool for guiding experimental MOF design. Full article

This paper presents the design and implementation of an autonomous robotic facility for textile sorting and recycling, leveraging advanced computer vision and machine learning technologies. The system enables real-time textile classification, localization, and sorting on a dynamically moving conveyor belt. A custom-designed pneumatic gripper is developed for versatile textile handling, optimizing autonomous picking and placing operations. Additionally, digital simulation techniques are utilized to refine robotic motion and enhance overall system reliability before real-world deployment. The multi-threaded architecture facilitates the concurrent and efficient execution of textile classification, robotic manipulation, and conveyor belt operations. Key contributions include (a) dynamic and real-time textile detection and localization, (b) the development and integration of a specialized robotic gripper, (c) real-time autonomous robotic picking from a moving conveyor, and (d) scalability in sorting operations for recycling automation across various industry scales. The system progressively incorporates enhancements, such as queuing management for continuous operation and multi-thread optimization. Advanced material detection techniques are also integrated to ensure compliance with the stringent performance requirements of industrial recycling applications. Full article

Artificial intelligence (AI) is reshaping mobile food-ordering apps, yet its impact on customer experience (CX) has not been fully mapped. Following systematic review guidelines (PRISMA 2020), a search of SCOPUS, Web of Science, ScienceDirect, and Google Scholar in March 2025 identified 55 studies published between 2022 and 2025. Since 2022, research has expanded from intention-based studies to include real-time app interactions and live app experiments. This shift has helped to identify five key CX dimensions: (1) instrumental usability: how quickly and smoothly users can order; (2) personalization value: AI-generated menus and meal suggestions; (3) affective engagement: emotional appeal of the app interface; (4) data trust and procedural fairness: users’ confidence in fair pricing and responsible data handling; (5) social co-experience: sharing orders and interacting through live reviews. Studies have shown that personalized recommendations and chatbots enhance relevance and enjoyment, while unclear surge pricing, repetitive menus, and algorithmic anxiety reduce trust and satisfaction. Given the limitations of this study, including its reliance on English-only sources, a cross-sectional design, and limited cultural representation, future research should investigate long-term usage patterns across diverse markets. This approach would help uncover nutritional biases, cultural variations, and sustained effects on customer experience. Full article

Face recognition systems often falter when deployed in uncontrolled settings, grappling with low light, unexpected occlusions, motion blur, and the degradation of sensor signals. Most contemporary algorithms chase raw accuracy yet overlook the pragmatic need for uncertainty estimation and multispectral reasoning rolled into a single framework. This study introduces HUE-Net—a Human-centric, Uncertainty-aware, Event-fused Network—designed specifically to thrive under severe environmental stress. HUE-Net marries the visible RGB band with near-infrared (NIR) imagery and high-temporal-event data through an early-fusion pipeline, proven more responsive than serial approaches. A custom hybrid backbone that couples convolutional networks with transformers keeps the model nimble enough for edge devices. Central to the architecture is the perturbed multi-branch variational module, which distills probabilistic identity embeddings while delivering calibrated confidence scores. Complementing this, an Adaptive Spectral Attention mechanism dynamically reweights each stream to amplify the most reliable facial features in real time. Unlike previous efforts that compartmentalize uncertainty handling, spectral blending, or computational thrift, HUE-Net unites all three in a lightweight package. Benchmarks on the IJB-C and N-SpectralFace datasets illustrate that the system not only secures state-of-the-art accuracy but also exhibits unmatched spectral robustness and reliable probability calibration. The results indicate that HUE-Net is well-positioned for forensic missions and humanitarian scenarios where trustworthy identification cannot be deferred. Full article

This study analyses the dynamic impact between winter wheat grains (‘Memory’ cultivar) and a flat metal surface under normal collisions. Four moisture levels (7%, 10%, 13% and 16%) and impact velocities from 1.0 to 4.5 m·s⁻¹ were chosen to reflect conditions in agricultural machinery. A custom test rig—comprising a transparent drop guide, a high-sensitivity piezoelectric force sensor and a high-speed camera—recorded grain velocity by vision techniques and contact force at 1 MHz. Force–time curves were examined to evaluate restitution velocity, the coefficient of restitution (CoR) and the effect of moisture on elastic–plastic deformation. CoR decreased non-linearly as impact velocity rose from 1.0 to 5.0 m·s⁻¹, and moisture content increased from 7% to 16%, falling from ≈ 0.60 to 0.40–0.50. Grains with higher moisture struck at higher velocities showed greater plastic deformation, longer contact times and intensified energy dissipation, making them more susceptible to internal damage. The data provide validated reference values for discrete element method (DEM) calibration and will assist engineers in designing grain-handling equipment that minimises mechanical damage during harvesting, conveying and processing. Full article

The controllable transport of droplets on solid surfaces is crucial for many applications, from water harvesting to bio-analysis. Herein, we propose a novel droplet transport controlling method, reconfigurable orbital electrowetting (ROEW) on inclined slippery liquid-infused porous surfaces (SLIPS), which enables controllable transport and dynamic handling of droplets by non-contact reconfiguration of orbital electrodes. The flexible reconfigurability is attributed to the non-contact wettability modulation and reversibly deformable flexible electrodes. ROEW graphically customizes stable wettability pathways by real-time and non-contact printing of charge-orbit patterns on SLIPS to support the continuous transport of droplets. Benefiting from the fast erase-writability of charges and the movability of non-contact electrodes, ROEW enables reconfiguration of the wetting pathways by designing electrode shapes and dynamically switching electrode configurations, achieving controllable transport of various pathways and dynamic handling of droplet sorting and mixing. ROEW provides a new approach for reconfigurable, electrode-free arrays and reusable microfluidics. Full article

In modern digital transactions involving government agencies, financial institutions, and commercial enterprises, reliable identity verification is essential to ensure security and trust. Traditional methods, such as submitting photocopies of ID cards, are increasingly susceptible to identity theft and fraud. To address these challenges, this study proposes a novel and robust identity verification framework that integrates super-resolution preprocessing, a convolutional neural network (CNN), and Monte Carlo dropout-based Bayesian uncertainty estimation for enhanced facial recognition in electronic know your customer (e-KYC) processes. The key contribution of this research lies in its ability to handle low-resolution and degraded facial images simulating real-world conditions where image quality is inconsistent while providing confidence-aware predictions to support transparent and risk-aware decision making. The proposed model is trained on facial images resized to 24 × 24 pixels, with a super-resolution module enhancing feature clarity prior to classification. By incorporating Monte Carlo dropout, the system estimates predictive uncertainty, addressing critical limitations of conventional black-box deep learning models. Experimental evaluations confirmed the effectiveness of the framework, achieving a classification accuracy of 99.7%, precision of 99.2%, recall of 99.3%, and an AUC score of 99.5% under standard testing conditions. The model also demonstrated strong robustness against noise and image blur, maintaining reliable performance even under challenging input conditions. In addition, the proposed system is designed to comply with international digital identity standards, including the Identity Assurance Level (IAL) and Authenticator Assurance Level (AAL), ensuring practical applicability in regulated environments. Overall, this research contributes a scalable, secure, and interpretable solution that advances the application of deep learning and uncertainty modeling in real-world e-KYC systems. Full article

This research addresses the challenge of precise trajectory tracking for cart–pendulum robotic systems affected by unknown nonlinear actuator dynamics. We introduce a novel control framework that combines neural network modeling with adaptive parameter estimation to handle these complex dynamics. By characterizing state-dependent actuator behavior through custom-designed linear filters and adaptive laws, our approach identifies system parameters with high precision. We then develop an innovative fixed-time adaptive sliding mode controller that guarantees convergence within a predetermined timeframe regardless of initial conditions. Lyapunov stability analysis confirms that tracking errors converge to a small neighborhood around zero within the specified time bounds, with the size of the neighborhood determined by the design parameters. Simulation studies on a watermelon transportation robot validate our approach’s practical effectiveness, demonstrating improved tracking accuracy and robustness against actuator disturbances compared with conventional methods. Full article

Background: The vitreomacular interface (VMI) encompasses a group of retinal disorders that significantly impact vision, requiring accurate classification for effective management. This study aims to compare the effectiveness of an expert-designed custom deep learning (DL) model and a code free Auto Machine Learning (ML) model in classifying optical coherence tomography (OCT) images of VMI disorders. Materials and Methods: A balanced dataset of OCT images across five classes—normal, epiretinal membrane (ERM), idiopathic full-thickness macular hole (FTMH), lamellar macular hole (LMH), and vitreomacular traction (VMT)—was used. The expert-designed model combined ResNet-50 and EfficientNet-B0 architectures with Monte Carlo cross-validation. The AutoML model was created on Google Vertex AI, which handled data processing, model selection, and hyperparameter tuning automatically. Performance was evaluated using average precision, precision, and recall metrics. Results: The expert-designed model achieved an overall balanced accuracy of 95.97% and a Matthews Correlation Coefficient (MCC) of 94.65%. Both models attained 100% precision and recall for normal cases. For FTMH, the expert model reached perfect precision and recall, while the AutoML model scored 97.8% average precision, and 97.4% recall. In VMT detection, the AutoML model showed 99.5% average precision with a slightly lower recall of 94.7% compared to the expert model’s 95%. For ERM, the expert model achieved 95% recall, while the AutoML model had higher precision at 93.9% but a lower recall of 79.5%. In LMH classification, the expert model exhibited 95% precision, compared to 72.3% for the AutoML model, with similar recall for both (88% and 87.2%, respectively). Conclusions: While the AutoML model demonstrated strong performance, the expert-designed model achieved superior accuracy across certain classes. AutoML platforms, although accessible to healthcare professionals, may require further advancements to match the performance of expert-designed models in clinical applications. Full article