Search Results (450)

Named Entity Recognition (NER) plays a crucial role in Aspect-Based Sentiment Identification (ABSI), enabling the extraction of domain-specific aspects and their associated sentiment expressions from unstructured textual reviews. In complex domains such as movie reviews, sentiment is frequently conveyed through references to named entities (e.g., actors, directors, or movie titles) and other contextual cues. However, many existing ABSI approaches treat NER as a separate preprocessing step, limiting the effective modeling of entity–aspect–opinion relationships. Integrating NER directly into the ABSI framework, allows entity-specific opinions to be more accurately identified, overlapping aspects to be disambiguated, and contextual sentiment expressions to be captured more effectively. To address these challenges, this study proposes an integrated NER-based aspect identification model built on feature-enhanced LSTM and BiLSTM architectures. Linguistic features, including Parts-of-Speech (POS) tags and chunking information, are incorporated to enrich contextual representations, while a Conditional Random Field (CRF) decoding layer models inter-label dependencies for coherent sequence-level predictions of named entities, aspects, and associated opinion expressions. Compared with large transformer-based models, the proposed BiLSTM-CRF architecture offers lower computational complexity, fewer parameters, and allows explicit integration and analysis of linguistic features that are often implicitly encoded in transformer attention mechanisms. The model is evaluated through multiple experimental variants across three domains. Four configurations are applied to movie-review data to jointly extract person names, movie titles, and aspect-opinion pairs, while six configurations assess cross-domain robustness on restaurant and laptop review datasets. Results show that the BiLSTM-CRF model augmented with POS features consistently outperforms baseline configurations in the movie domain and remains competitive across domains, achieving an F1-score of 0.89. These findings demonstrate that explicit linguistic feature integration within a CRF-based sequence modeling can provide an effective and computationally efficient alternative to large-scale transformer fine-tuning for structured, entity-linked ABSI tasks. Full article

Federated learning (FL) enables collaborative model training across distributed devices and robotic systems while preserving data privacy, making it well-suited for swarm robotics and edge-device-powered intelligence. However, FL remains vulnerable to adversarial behaviors such as data and model poisoning, particularly in real-world deployments where detection methods must operate under strict computational and communication constraints. This paper presents a practical, real-world federated learning framework that enhances robustness to adversarial agents in a ROS2-based edge-device swarm environment. The proposed system integrates the Federated Averaging (FedAvg) algorithm with a lightweight average cosine similarity-based filtering method to detect and suppress harmful model updates during aggregation. Unlike prior work that primarily evaluates poisoning defenses in simulated environments, this framework is implemented and evaluated on physical hardware, consisting of a laptop-based aggregator and multiple Raspberry Pi worker nodes. A convolutional neural network (CNN) based on the MobileNetV3-Small architecture is trained on the MNIST dataset, with one worker executing a sign-flipping model poisoning attack. Experimental results show that FedAvg alone fails to maintain meaningful model accuracy under adversarial conditions, resulting in near-random classification performance with a final global model accuracy of 11% and a loss of 2.3. In contrast, the integration of cosine similarity filtering demonstrates effective detection of sign-flipping model poisoning in the evaluated ROS2 swarm experiment, allowing the global model to maintain model accuracy of around 90% and loss around 0.37, which is close to baseline accuracy of 93% of the FedAvg algorithm only under no attack with a very minimal increase in loss, despite the presence of an attacker. The proposed method also maintains a false positive rate (FPR) of around 0.01 and a false negative rate (FNR) of around 0.10 of the global model in the presence of an attacker, which is a minimal difference from the baseline FedAvg-only results of around 0.008 for FPR and 0.07 for FNR. Additionally, the proposed method of FedAvg + cosine similarity filtering maintains computational statistics similar to baseline FedAvg with no attacker. Baseline results show an average runtime of about 34 min, while our proposed method shows an average runtime of about 35 min. Also, the average size of the global model being shared among workers remains consistent at around 7.15 megabytes, showing little to no increase in message payload sizes between baseline results and our proposed method. These results demonstrate that computationally lightweight cosine similarity-based detection methods can be effectively deployed in real-world, resource-constrained robotic swarm environments, providing a practical path toward improving robustness in real-world federated learning deployments beyond simulation-based evaluation. Full article

High-frequency non-intrusive load monitoring provides detailed harmonic information for appliances’ power disaggregation, and machine-learning approaches have demonstrated good performance in this task. However, these methods provide little transparency regarding the information structure of the aggregate signal. To address this, this paper models NILM as a coding-decoding process and applies information-theoretic measures to quantify uncertainty, recoverability, temporal contribution, and inter-appliance masking effects in aggregate signals. In the analyzed dataset, transfer entropy suggests negligible temporal gains, which is consistent with the observed effectiveness of pointwise models such as Random Forest. Moreover, conditional mutual information emphasizes the asymmetric masking relationships between appliances, with the laptop charger acting as a dominant interferer in the considered measurements. These findings are validated through a Random Forest regression model with minimum Redundancy Maximum Relevance feature selection. The results show that the mutual information between an appliance and the aggregate is a good predictor of disaggregation performance in the examined data, as appliances with high mutual information, such as hair dryer and electric water heater, achieve lower estimation errors, while others, such as iron, are difficult to recover despite stable distributions. This relationship is statistically supported by a strong negative monotonic correlation between normalized mutual information and the disaggregation error (Spearman $r_s=-0.81$, $p=0.015$). Hence, this work demonstrates how information-theoretic analysis can help characterize disaggregation difficulty prior to model training and assess the observability of appliances in high-frequency NILM. Full article

Aspect-based sentiment analysis (ABSA) plays a vital role in deriving fine-grained sentiment from textual content. As large language models (LLMs) are increasingly adopted for automated data annotation in natural language processing (NLP), concerns have emerged regarding the accuracy of their outputs. Despite their capacity to generate large volumes of labeled data, LLMs often suffer from overconfidence in predictions, high uncertainty in complex contexts, and difficulty capturing nuanced meanings, which compromise the quality of annotations and, in turn, the performance of downstream models. This underscores the need to enhance LLM adaptability while maintaining annotation accuracy. To address these limitations, we integrated low-rank adaptation (LoRA) with parameter-efficient fine-tuning (PEFT) for adapting Qwen2.5 to ABSA. LoRA reduces the number of trainable parameters by decomposing weight updates into low-rank matrices, while PEFT introduces modular adapter layers with scaled gradient updates and dynamic rank allocation. Using the standard SemEval 2014 Laptop dataset, Qwen2.5-3B fine-tuned with LoRA and PEFT achieves 64.50% accuracy, outperforming its baseline of 24.05%. Likewise, Qwen2.5-7B attains 77.50%, compared with a baseline of 34.63%. These results highlight the potential of parameter-efficient methods to improve the accuracy of LLMs in ABSA annotation tasks, especially under resource constraints. Such results lay the groundwork for scalable, reproducible LLM deployment and open avenues for future research in cross-domain adapter transferability and dynamic rank optimization. Full article

Injection-related errors remain a common clinical issue and can cause patient discomfort, hematoma formation, and procedural inefficiencies. The visualization of subcutaneous veins using near-infrared (NIR) imaging has gained attention as an effective approach to reducing such errors, as blood exhibits a higher absorption of NIR light than surrounding tissue. In this study, a low-cost, non-invasive vein visualization system is presented to support safer and more accurate venous access. The proposed system integrates an NIR illumination source and a modified webcam within a compact equipment enclosure, allowing subjects to be conveniently examined by placing their arm inside the device. Vein images are automatically acquired using a laptop-based platform, followed by digital image processing techniques for vein enhancement and visualization. Laboratory-scale experiments were conducted on healthy volunteers to evaluate system performance under multiple conditions, including different vein locations (upper and lower arm regions), varying distances between the NIR light source and the arm (15 cm and 20 cm), and ambient illumination interference (light sources on and off). The experimental results demonstrate the successful implementation and reliable operation of the proposed system. Effective vein visualization was achieved across all test conditions, as confirmed by qualitative visual assessment and quantitative image quality metrics, including the Peak Signal-to-Noise Ratio (PSNR) and Mean Squared Error (MSE). Overall, the proposed system offers a practical, accessible, and cost-effective solution for vein visualization, showing strong potential for clinical and experimental applications aimed at reducing injection errors and improving venous access reliability. Full article

Fully Homomorphic Encryption (FHE) provides strong data confidentiality during computation but often suffers from high latency on Central Processing Units (CPUs). This study evaluates Graphics Processing Unit (GPU) acceleration for modern FHE libraries across a laptop (NVIDIA GTX 1650 Ti), a server (NVIDIA RTX 4060), and a Jetson Nano 2 GB embedded GPU. We benchmark key generation, arithmetic operations, Boolean-gate evaluation and scheme-specific tasks such as relinearization and key switching, using library-provided benchmarks with an explicit baseline (operation scope, timing boundaries, and parameter tuples). Moreover, we compare GPU-native libraries (NuFHE, Phantom-FHE, and Troy-Nova) with CPU-oriented ones (Microsoft SEAL, HElib, OpenFHE, Cupcake, and TFHE-rs). Results show GPUs deliver significant speedups for targeted operations. For example, NuFHE’s NVIDIA CUDA (Compute Unified Device Architecture) backend achieves about 1.4× faster Boolean-gate evaluation on the laptop and 3.4× faster on the server compared to its OpenCL backend. Likewise, RLWE (Ring Learning With Errors)-based schemes (BFV, CKKS, and BGV) see marked gains for polynomial arithmetic such as Number Theoretic Transform (NTT) when executed via Phantom-FHE. However, attempts to add CUDA support to Microsoft SEAL reveal four main challenges: high-precision modular arithmetic on GPUs, sequential dependencies in SEAL’s design, limited GPU memory and complex build-system changes. In light of these findings, we propose revised guidelines for GPU-first FHE libraries and practical recommendations for deploying high-throughput, privacy-preserving solutions on modern GPUs. Full article

We report a chromosome-level assembly of a male red coachwhip snake (Masticophis flagellum piceus) generated exclusively with nanopore sequencing. Using Hifiasm-ONT for assembly and RagTag for scaffold polishing, we produced a 1.61 Gb nuclear genome comprising 8 macrochromosomes and 10 microchromosomes with a 97.7% BUSCO completeness score. Annotation with LiftOn found 19,832 loci, including 18,025 protein-coding genes. The mitochondrial genome, assembled with MitoHiFi and annotated with MitoFinder, was 17,119 bp with 13 coding genes, 22 tRNAs and 2 rRNAs. All sequencing was performed in a simulated mobile laboratory using a portable sequencer and a laptop with analyses done both locally and remotely. These results highlight the feasibility of decentralized genomics and its potential to accelerate biodiversity research globally. Full article

Interactive teleoperation offers an intuitive pathway for human–robot interaction, yet many existing systems rely on dedicated sensors or wearable devices, limiting accessibility and scalability. This paper presents a vision-based teleoperation framework that enables real-time control of an articulated robotic arm (five joints plus a gripper actuator) using human hand tracking from a single, typical laptop camera. Hand pose and gesture information are extracted using a real-time landmark estimation pipeline, and a set of compact kinematic descriptors—palm position, apparent hand scale, wrist rotation, hand pitch, and pinch gesture—are mapped to robotic joint commands through a calibration-based control strategy. Commands are transmitted over a lightweight network interface to an embedded controller that executes synchronized servo actuation. To enhance stability and usability, temporal smoothing and rate-limited updates are employed to mitigate jitter while preserving responsiveness. In a human-in-the-loop evaluation with 42 participants, the system achieved an 88% success rate (37/42), with a completion time of 53.48 ± 18.51 s, a placement error of 6.73 ± 3.11 cm for successful trials (n = 37), and an ease-of-use score of 2.67 ± 1.20 on a 1–5 scale. Results indicate that the proposed approach enables feasible interactive teleoperation without specialized hardware, supporting its potential as a low-cost platform for robotic manipulation, education, and rapid prototyping. Full article

In this work, we use an Echo State Network (ESN) model, which is essentially a recurrent neural network (RNN) operating according to the reservoir computing (RC) paradigm, to classify individual ECG heartbeats using the MIT-BIH arrhythmia database. The aim is to evaluate the performance of ESN in a challenging task that involves classification of complex, unprocessed one-dimensional signals, distributed into five classes. Moreover, we investigate the performance of the ESN in the presence of (i) noise in the dynamics of the internal variables of the hidden (reservoir) layer and (ii) random variability in the activation functions of the hidden layer cells (neurons). The overall accuracy of the best-performing ESN, without noise and variability, exceeded 96% with per-class accuracies ranging from 90.2% to 99.1%, which is higher than previous studies using CNNs and more complex machine learning approaches. The top-performing ESN required only 40 min of training on a CPU (Intel i5-1235U@1.3 GHz) HP laptop. Notably, an alternative ESN configuration that matched the accuracy of a prior CNN-based study (93.4%) required only 6 min of training, whereas a CNN would typically require an estimated training time of 2–3 days. Surprisingly, ESN performance proved to be very robust when Gaussian noise was added to the dynamics of the reservoir hidden variables, even for high noise amplitudes. Moreover, the success rates remained essentially the same when random variability was imposed in the activation functions of the hidden layer cells. The stability of ESN performance under noisy conditions and random variability in the hidden layer (reservoir) cells demonstrates the potential of analog hardware implementations of ESNs to be robust in time-series classification tasks. Full article

This paper reports on a fully automatic structured Computational Fluid Dynamics (CFD) mesh generation method based on a parametric multi-block topology for airfoil configurations. The method parameterizes the control vertices and the control edges, which construct the multi-block topology, and then assembles blocks with the control edges. Once the airfoil shape is determined, the topology is immediately updated based on the parameterization, and the CFD mesh of the airfoil is generated using transfinite interpolation. The present method is tested on airfoils with different topologies, shapes, and mesh sizes to check its robustness, efficiency, and quality. The test results show that the mesh of an airfoil of any shape can be generated automatically with high quality. In addition, an airfoil CFD mesh with about 50 million nodes can be automatically generated in less than ten seconds on a laptop, and the Jacobi of over 97% of the mesh cells is larger than 0.9. The flow simulation results for the NACA0012 airfoil agree well with the wind-tunnel test data, demonstrating the method’s applicability to CFD. Full article

Several commercial seismometers now support CSTP, the real-time communication protocol used in the China Earthquake Early Warning System, but there is still no simple, flexible, and low-cost solution to archive CSTP streams or integrate them into existing data processing systems. In this study, we design and implement SharpCEEWPServer, a lightweight, out-of-the-box graphical server that integrates client management, real-time data reception, visualization, and archiving, and can, via RingServer, convert CSTP real-time streams into widely supported SeedLink streams. Hardware compatibility is evaluated using four commercial CSTP-capable instruments, a forwarding chain is built to assess forwarding functionality and reliability, and concurrency performance is tested using simulated networks with different station counts. The stability under network impairment scenarios and the performance of the forwarding system were also analyzed. The results show that the server can reliably receive and forward real-time data streams, and that laptop-class hardware is sufficient to withstand the load imposed by an M7.0 earthquake scenario when receiving real-time streams from 1000 three-component seismometers. However, the current version’s latency performance can only meet the needs of non-early warning networks. Overall, the proposed server significantly lowers the deployment and usage threshold for new CSTP-capable instruments and provides an efficient, low-cost integration solution for temporary networks in earthquake emergency response and seismic arrays. Full article

Background: Decoding covert spatial attention (CSA) from dry, low-channel electroencephalography (EEG) is key for gaze-independent brain–computer interfaces (BCIs). Methods: We evaluate, on sixteen participants and three tasks (CSA, motor imagery (MI), Emotion), a four-electrode, subject-wise pipeline combining leak-safe preprocessing, multiresolution wavelets, and a compact Hybrid encoder (CNN-LSTM-MHSA) with robustness-oriented training (noise/shift/channel-dropout and supervised consistency). Results: Online, the Hybrid All-on-Wav achieved 0.695 accuracy with end-to-end latency ~2.03 s per 2.0 s decision window; the pure model inference latency is ≈185 ms on CPU and ≈11 ms on GPU. The same backbone without defenses reached 0.673, a CNN-LSTM 0.612, and a compact CNN 0.578. Offline subject-wise analyses showed a CSA median Δ balanced accuracy (BAcc) of +2.9%p (paired Wilcoxon p = 0.037; N = 16), with usability-aligned improvements (error 0.272 → 0.268; information transfer rate (ITR) 3.120 → 3.240). Effects were smaller for MI and present for Emotion. Conclusions: Even with simple hardware, compact attention-augmented models and training-time defenses support feasible, low-latency left–right CSA control above chance, suitable for embedded or laptop-class deployment. Full article

A person’s face can reveal their mood, and microexpressions, although brief and involuntary, are also authentic. People can recognize facial gestures; however, their accuracy is inconsistent, highlighting the importance of objective computational models. Various artificial intelligence models have classified microexpressions into three categories: positive, negative, and surprise. However, it is still significant to address the basic Ekman microexpressions (joy, sadness, fear, disgust, anger, and surprise). This study proposes a Transformers-based machine learning model, trained on CASME, SAMM, SMIC, and its own datasets. The model offers comparable results with other studies when working with seven classes. It applies various component-based techniques ranging from ViT to optical flow with a different perspective, with low training rates and competitive metrics comparable with other publications on a laptop. These results can serve as a basis for future research. Full article

The ergonomic risks associated with posture in conventional office workstations have been extensively studied, but there is limited research available on these risks in the context of home-based work environments. Most available studies rely solely on questionnaire-based statistical analyses, leaving a gap in understanding the specific conditions of home-based work environments. This study focuses on evaluating the effects of workstation conditions on posture and muscular efforts across three anatomical segments: head-neck, trunk-upper trapezius, and arm-deltoid. The analysis is conducted by simulating workstation setups commonly associated with academic activities performed by students during the COVID-19 pandemic. The conditions examined in this study include inadequate desk height, the use of chairs without armrests, and the use of laptops. Eighteen volunteers, comprising nine women and nine men, participated in experiments conducted under scenarios designed using a $2^k$ statistical approach. In all experiments, participants completed questionnaires, and text-writing activities were performed to evaluate the effects of these conditions. This research introduces a new non-invasive technique for ergonomic assessment that integrates photogrammetry and surface electromyography (sEMG) to simultaneously evaluate posture and muscular effort. The developed methodology allows precise, contactless analysis of ergonomic conditions and can be adapted for various professional and academic teleworking environments. Significant effects were observed in the posture (°) of the trunk and head, with both small and large effects identified at significance levels of p < 0.001 under the furniture conditions studied. In terms of EMG activity, moderate effects were observed at p < 0.01 levels between table height and upper trapezius activation, while small effects were detected at p < 0.05 levels between the use of chairs without armrests and neck. Similarly, small to moderate effects were observed in the arm-deltoid segment under the same furniture conditions. These findings reveal information about the posture and muscular effort patterns associated with the studied tasks, offering knowledge that can be referenced for similar tasks in other technical fields where telematics activities are performed. Full article

The recovery of critical metals from spent lithium-ion batteries (LIBs) is essential to reduce environmental impacts and promote circular economy strategies. This study developed a sustainable and scalable process for the recovery and complete valorization of lithium, cobalt, and other valuable components from end-of-life LIBs. Hydrometallurgical treatment using biodegradable citric and oxalic acids was employed as a green alternative to conventional inorganic acids, achieving high selectivity and reduced environmental impact. Experimental work was conducted on 3 kg of LIBs from discarded laptop batteries (Dell and HP). After safe discharge and dismantling, the cathode materials were thermally treated at 300 °C to detach active components, followed by acid leaching in 1 M citric acid at 30 °C, pH 2.5, and 6 h of reaction. Lithium and cobalt were recovered as oxalates with efficiencies of 90% and 85%, respectively, while copper, aluminum, and graphite were separated through mechanical and thermal processes. Beyond metal recovery, the process demonstrates a circular upcycling approach, transforming recovered materials into functional products such as aluminum keychains, copper jewelry, and graphite-based pencils. This integrated strategy connects hydrometallurgical extraction with material reuse, advancing toward a zero-waste, closed-loop system for sustainable LIB recycling and local resource valorization. Full article