IEEE Access最新文献

{"title":"Discovering Domain-Agnostic Fake News Detectors Through Deep Self-Supervised Learning","authors":"Carmela Comito;Massimo Guarascio;Angelica Liguori;Giuseppe Manco;Francesco Sergio Pisani","doi":"10.1109/ACCESS.2025.3608790","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3608790","url":null,"abstract":"The rapid spread of misinformation across online platforms poses a major threat to societal trust, public health, and democratic processes. While recent advances in machine learning have improved the accuracy of fake news detection, most existing approaches remain limited to single-domain settings and struggle to generalize across diverse domains or platforms. To address this challenge, we propose <italic>DAFNE</i> (<bold>D</b>omain-<bold>A</b>gnostic <bold>F</b>ake <bold>NE</b>ws detector), a deep learning approach designed to capture cross-domain high-level features for fake news detection. By combining feature-level adversarial learning with self-supervised learning, <italic>DAFNE</i> effectively learns domain-invariant representations that enable reliable detection across heterogeneous sources. The proposed approach is evaluated on five real-world benchmark datasets spanning multiple domains, and the results demonstrate superior generalization capabilities compared to state-of-the-art baselines. Specifically, <italic>DAFNE</i> outperforms the competitors, with average micro-F1 improvements ranging from 11.3% to 39.9%. In comparison to the second-best model, our approach shows an average improvement of 18% across all domains in terms of the F-Score, reaching up to 25% on the Politifact dataset. These results highlight the capability of <italic>DAFNE</i> to mitigate the domain shift problem, enabling more reliable and adaptive misinformation detection in dynamic online environments.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"147408-147421"},"PeriodicalIF":3.6,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11159185","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145089973","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Real-Time Detection of Skin-Electrode Adhesion Based on Embedded Neural Networks for Bioimpedance Spectroscopy","authors":"Rosanna Manzo;Andrea Apicella;Pasquale Arpaia;Francesco Caputo;and Nicola Moccaldi","doi":"10.1109/ACCESS.2025.3605928","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3605928","url":null,"abstract":"A module based on embedded Multi-Layer Perceptrons (MLPs) was developed for real-time monitoring of skin-electrode adhesion quality. It was designed to integrate with Insulin-Meter, an established 4-wire bioimpedance spectroscopy system for measuring insulin absorption in diabetic patients, reported in previous studies. The MLPs address two classification tasks in cascade: (i) adhesion vs. partial detachment and (ii) identification of the partially detached electrode. The MLPs can be deployed on the same microcontroller used for insulin absorption assessment, leveraging the same impedance spectroscopy data. In literature, adhesion monitoring based on impedance measurement has been implemented in applications with unfavorable signal-to-noise ratio (SNR), such as electroencephalography (EEG), where contact quality is typically verified prior to signal acquisition using threshold-based approach. For other biosignal measurements, the higher signal-to-noise ratio and shorter acquisition durations have generally made real-time monitoring of electrode-skin adhesion unnecessary. However, Insulin-Meter requires extended acquisition sessions under unfavorable SNR conditions. MLPs were compared to other machine learning algorithms, including Logistic Regression, Support Vector Machines and Random Forest. Hyperparameter optimization was performed with consideration for the memory footprint of all classifiers. The MLPs outperformed the other algorithms and were deployed on a low-cost, general-purpose microcontroller, requiring significantly less than 50 % of its flash memory. The system achieved an accuracy of 98 % <inline-formula> <tex-math>$pm ~3$ </tex-math></inline-formula> % for discriminating between adhesion and partial detachment, and 97 % <inline-formula> <tex-math>$pm ~13$ </tex-math></inline-formula> % for identifying the partially detached electrode. The microcontroller requires an average inference time of 4.286 ms to implement the two-step classification task.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"155385-155398"},"PeriodicalIF":3.6,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11151666","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145021151","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Next-Gen Aviation: Who Will Rule the Skies—Hydrogen or Electric?—A Review","authors":"Saikat Chowdhury;Mona Ghassemi","doi":"10.1109/ACCESS.2025.3605809","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3605809","url":null,"abstract":"In pursuit of net-zero aviation by 2050, fully electric and hydrogen-fueled aircraft have emerged as two leading pathways to decarbonize the aviation sector. This study examines the pathways of fully electric and hydrogen-fueled aircraft as potential solutions for achieving net-zero aviation by 2050. Through a systematic literature review, we evaluate the energy efficiency, storage requirements, infrastructure needs, and environmental impacts associated with each technology. Our findings indicate that electric aircraft offer immediate advantages for short-haul operations, including reduced operating costs and compatibility with existing infrastructure; however, the low energy density of batteries currently limits their potential. Conversely, hydrogen propulsion holds promise for long-range flights due to its high energy content, but it necessitates advancements in liquid hydrogen storage, distribution, and green hydrogen production for practical implementation. This analysis offers valuable insights for industry stakeholders and policymakers aiming to advance sustainable aviation technologies.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"155141-155154"},"PeriodicalIF":3.6,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11150403","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145021250","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Context-Aware Directed Acyclic Graph Network for Conversational Aspect-Based Sentiment Quadruple Analysis","authors":"Qiang Zhang;Jie Zeng;Runze Zhang;Dong Cui","doi":"10.1109/ACCESS.2025.3605729","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3605729","url":null,"abstract":"Conversational Aspect-based Sentiment Quadruple Analysis (DiaASQ) is a fine-grained sentiment analysis task that aims at extracting targets, aspects, opinions, and sentiments from multi-turn dialogues. Existing methods focus on token-level interaction modeling and neglect complex cross-utterance dependencies. To address this, we propose a context-aware directed acyclic graph network (CA-DAGNet). This model integrates syntax-aware context encoding and directed acyclic graph (DAG) modeling to capture intra-utterance syntactic structures and cross-utterance long-range dependencies. For global modeling, we construct the dialogue as a DAG and combine it with an information propagation mechanism, precisely capturing syntactic dependencies and semantic interactions while dynamically adjusting the scope of information propagation to avoid fixed-window limitations. In addition, we adopt a context filter to retain highly relevant information for the target utterance, suppress redundant noise, and improve the modeling of cross-utterance dependencies. Experiments conducted on Chinese and English datasets demonstrate that the proposed model achieves superior performance.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"154823-154832"},"PeriodicalIF":3.6,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11148487","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145021276","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Highly Sensitive Optic Fabry-Perot Ultrasonic Sensor for Power Transformer Partial Discharge Detection","authors":"Jinggang Yang;Qun Li;Jiabi Liang;Jian Shao;Peng Wu;Tonglei Wang;Yuncai Lu;Xiaohan Li","doi":"10.1109/ACCESS.2025.3605613","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3605613","url":null,"abstract":"Optical sensing has been widely applied in the condition monitoring of power equipment due to its advantages, such as high insulation, anti-interference capability, and high sensitivity. However, for partial discharge detection, the sensitivity and applicability of optical sensors still need to be further improved to ensure their practical application. This paper introduces a silicon-grooved diaphragm-based fiber-optic Fabry-Perot (F-P) ultrasonic sensor that was designed and fabricated for partial discharge detection in power equipment such as power transformers. The groove parameters of the sensing diaphragm were optimized using finite element software. Compared to traditional circular diaphragms, the static sensitivity of the silicon-grooved diaphragm was improved by 4.09 times, while the resonant frequency remained essentially unchanged. The influence of the F-P cavity length on the contrast of the sensor’s interference spectrum was investigated by coupling efficiency to modify the traditional dual-beam interference model, thereby enhancing the sensor’s acoustic pressure sensitivity. The silicon grooved diaphragm was fabricated using micro-electro-mechanical system (MEMS) technology, with a groove diameter of <inline-formula> <tex-math>$829.44~mu $ </tex-math></inline-formula>m, a thickness of <inline-formula> <tex-math>$2.09~mu $ </tex-math></inline-formula>m, and an F-P cavity length of <inline-formula> <tex-math>$163.600~mu $ </tex-math></inline-formula>m. At the resonant frequency of 61.5 kHz, the sensor achieved an acoustic pressure sensitivity of 357.78 mV/Pa. The performance of the sensor was validated by testing in a power transformer with three types of partial discharge defect models. Experimental results demonstrate that the fabricated fiber-optic F-P ultrasonic sensor offers high acoustic pressure sensitivity, good real-time performance, and capabilities in detecting ultrasonic signals. In addition, the developed sensor maintains structural integrity and can function after long-term usage in a transformer environment.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"154898-154907"},"PeriodicalIF":3.6,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11149660","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145021401","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Observer-Based Induced L∞ Control for Nonlinear Systems With Deception Attacks","authors":"Ying Tian;Ming-Yang Qiao","doi":"10.1109/ACCESS.2025.3605687","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3605687","url":null,"abstract":"In this paper, the observer-based induced <inline-formula> <tex-math>${mathcal {L}}_{infty }$ </tex-math></inline-formula> control problem for nonlinear systems with deception attacks is studied. Construct the T–S fuzzy model to linearly approximate the nonlinearity of the system. Randomly occurring deception attack is assumed obey the Bernoulli distribution and the norm of the attack function satisfies certain conditions. Furthermore, the closed-loop control system (CLCS) is constructed in the descriptor form to directly separate the system matrices and the controller matrix in order to reduce the design complexity. The proposed design conditions enable the observer and controller to meet the requirements of stochastic stability and induced <inline-formula> <tex-math>${mathcal {L}}_{infty }$ </tex-math></inline-formula> performance. Finally, the algorithm was verified through a one-link manipulator.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"154976-154983"},"PeriodicalIF":3.6,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11150419","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145021245","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AnnotationGym: A Generic Framework for Automatic Source Code Annotation","authors":"Hafsah Shahzad;Ahmed Sanaullah;Sanjay Arora;Ulrich Drepper;Martin C. Herbordt","doi":"10.1109/ACCESS.2025.3605852","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3605852","url":null,"abstract":"A common approach to code optimization is to insert compiler hints in the source code using annotations. Two major challenges with using annotations effectively are their complexity and lack of portability. This means, first, that significant developer expertise is required, and, second, that the supported annotations, as well as their syntax and use, can vary substantially. Moreover, there is not currently any tool that can output performant annotation-inserted codes for different back-ends. To address these challenges, we present AnnotationGym, an easy-to-use, open-source, generic infrastructure that supplements or replaces the developer in annotating source code. It demonstrates a novel application of AI methods to code annotation. In addition to improving code performance, the flexibility of AnnotationGym enables easy comparisons of performance and optimization strategies among compilers and target architectures and thus provides an extensible platform to facilitate further progress in this field. AnnotationGym automatically extracts structured information about the target code and compiler to generate a list of possible annotations. AI-based optimization algorithms then traverse this space to determine the best set of annotations depending on the developer goals. To demonstrate its effectiveness, we run AnnotationGym on popular, representative workloads from the Polybench suite, as well as targeting various compilers (GCC, AMD HLS, Intel HLS), optimization algorithms (Reinforcement Learning, Bayesian Optimization), and architectures (CPU, FPGA). We also test our approach on FPGA codes derived, e.g., from the Rodinia and OpenDwarfs benchmarks and that are hand-optimized using standard best practices. An interesting finding is that the best overall performance obtained by AnnotationGym was generally with unoptimized codes.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"155321-155339"},"PeriodicalIF":3.6,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11148243","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145021200","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Magnetic Continuum Robots for the Treatment of Atrial Fibrillation","authors":"Yuqing Tang","doi":"10.1109/ACCESS.2025.3605811","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3605811","url":null,"abstract":"Atrial fibrillation (AF), the most common sustained cardiac arrhythmia, imposes a major global health burden. While antiarrhythmic drugs remain first-line therapy for rhythm control, their long-term efficacy is limited by variable response rates, adverse effects, and high recurrence rates in persistent AF. Radiofrequency catheter ablation (RFCA), primarily targeting pulmonary vein isolation (PVI), has emerged as a corner-stone intervention for AAD-refractory patients. However, conventional catheters face limitations in navigating complex anatomy and maintaining tissue contact, impacting efficacy and safety. To overcome these challenges and improve outcomes for drug-resistant AF, magnetic navigation systems (MNS) offer enhanced precision. Integrating MNS with robotic platforms and force sensing unlocks significant potential. This review examines Magnetic Continuum Robots (MCRs) with force feedback for AF ablation. We detail MNS principles, MCR design/actuation, and the role of force sensing in optimizing lesion formation—critical for durable PVI and reducing AF recurrence post-ablation. Pre-clinical and clinical data demonstrate advantages including navigation accuracy and reduced complications. MCRs show potential to address pharmacological limitations by offering a promising interventional approach for AAD-refractory AF, though further clinical validation is required.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"155355-155366"},"PeriodicalIF":3.6,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11148486","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145021229","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Trajectory and Parameter Optimization in Robust Tracking Control of a Quadrotor","authors":"Ngoc-Hiep Tran;Quy-Thinh Dao;Thi-van-Anh Nguyen;Ngoc-Tam Bui","doi":"10.1109/ACCESS.2025.3605761","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3605761","url":null,"abstract":"Research on combined control methods for quadrotors has focused on trajectory tracking, robust control, neural networks, parameter optimization, and path planning. While previous studies have not fully integrated all of these aspects, this study presents a comprehensive control framework that seamlessly combines robust control strategies, neural network-based uncertainty approximation, path planning, and optimization to achieve precise and reliable trajectory tracking of quadrotors operating under significant model uncertainties and external disturbances. At the heart of the framework is the Integrated sliding mode control (Intergrated SMC), a design that merges the inherent robustness of sliding mode control with the adaptive approximation capability of radial basis function (RBF) neural networks. The fusion of these two elements not only ensures stability but also strengthens the system’s resilience, delivering high-precision tracking even in the presence of unmodeled dynamics and external disturbances. The framework also incorporates the rapidly-exploring random tree star (RRT*) algorithm for trajectory planning, allowing the generation of collision-free and asymptotically optimal reference paths capable of navigating environments with complex obstacle distributions. In addition, particle swarm optimization (PSO) is employed to systematically tune the controller gains and neural network parameters, thereby enhancing overall control performance. Extensive simulations under varying conditions of model mismatch and disturbances confirm the superior performance of the proposed integrated approach, demonstrating significant improvements in tracking accuracy and disturbance rejection compared to conventional control methods. This unified architecture thus provides a robust and computationally efficient solution for quadrotor trajectory tracking, maintaining high performance even in the presence of model uncertainties and external disturbances.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"155215-155232"},"PeriodicalIF":3.6,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11148240","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145021262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Recognizing Multigenerational Families From Videos in the Wild","authors":"Xiaoqian Qin;Bin Gui;Dong Wang","doi":"10.1109/ACCESS.2025.3605630","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3605630","url":null,"abstract":"In the field of computer vision, current relative learning research has predominantly concentrated on identifying parent-child relationships from pairs of static facial images, neglecting the discriminative information inherent in relatives involving multiple subjects and disregarding the common noise in unconstrained settings. To tackle these limitations, we present a novel task of video-based multigenerational family recognition, aiming to recognize multigenerational families from videos captured in unconstrained environments. We propose a Support Vector Data Description (SVDD)-based family-shared multi-metric learning (SFM2L) method, where only purified samples are subjected to multi-metric learning to derive both family-shared and family-specific distance metrics. To further improve the recognition performance, we introduce a multi-view method named MSFM2L, which effectively integrates deep and shallow features. In addition, we have constructed a new video dataset consisting of 90 multigenerational families. Extensive experiments on both the newly collected dataset and the well-established KinFaceW kinship face dataset clearly demonstrate the superior performance of our proposed methods compared to existing metric learning approaches.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"155233-155246"},"PeriodicalIF":3.6,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11148236","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145021323","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}