计算机科学最新文献

{"title":"Application of differential privacy in smart building systems","authors":"Yuanxiu Teng ,&nbsp;Li Yin ,&nbsp;Yufeng Chen ,&nbsp;Zhiwu Li ,&nbsp;Xin Hu ,&nbsp;Ahmed M. El-Sherbeeny","doi":"10.1016/j.sysconle.2025.106144","DOIUrl":"10.1016/j.sysconle.2025.106144","url":null,"abstract":"<div><div>Smart buildings are an important component of urban intelligence due to their efficient information management and engineering control. The large amount of data stored and shared in smart building systems may expose the privacy of users. This work deals with initial resource configuration information protection in a probabilistic automaton framework, using state-based differential privacy. For given two adjacent initial states, a verification method for state-based differential privacy is provided to verify whether an attacker can identify with which state the system is initialized. An enforcement method is proposed to ensure that two systems with adjacent initial states can generate observations with similar probability distributions. The examples in this paper show that applying differential privacy in smart building systems can effectively protect initial state information.</div></div>","PeriodicalId":49450,"journal":{"name":"Systems & Control Letters","volume":"203 ","pages":"Article 106144"},"PeriodicalIF":2.1,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144196376","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Temporal Neighbor Sequence-based Interpretable Spammer Groups Detection on E-commerce platform","authors":"Ning Li ,&nbsp;Shujuan Ji ,&nbsp;Yingtong Dou ,&nbsp;Dickson K.W. Chiu ,&nbsp;Qi Zhang ,&nbsp;Yongquan Liang ,&nbsp;Yongshan Wei","doi":"10.1016/j.ipm.2025.104177","DOIUrl":"10.1016/j.ipm.2025.104177","url":null,"abstract":"<div><div>Organized spammer groups collaborate to manipulate reviews for illicit gains, posing significant challenges to online platforms. This paper introduces a Temporal Neighbor Sequence-based Interpretable Spammer Group Detection method called TNSGD. First, we filter high-suspicious reviewers to reduce node complexity in the co-review temporal network, optimizing the detection process. Second, a co-review temporal network is constructed using these filtered reviewers, generating temporal neighbor sequences that capture temporal aggregation and relational features to form candidate groups. These candidate groups are then classified using group spam indicators and heuristic conditions to delineate the final spammer groups. TNSGD surpasses baseline methods with notable improvements in Precision and F1 scores, including enhancements of 4% and 3% for Amazon and 39% and 31% for Yelp, respectively. Additionally, TNSGD significantly reduces computational complexity to 1/85th and 1/7th. Furthermore, we provide interpretations of TNSGD from two perspectives: model and result. We devise a transparent detection process for model interpretation to ensure each step has a clear physical significance. For result interpretation, we offer interpretable visualizations of the temporal–spatial and evolutionary characteristics of the detected spammer groups, providing valuable insights for refining future detection models.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"62 6","pages":"Article 104177"},"PeriodicalIF":7.4,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144195258","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Conversion of bound states to noise-like pulses in an L-band mode-locked fiber laser","authors":"Qimeng Lin ,&nbsp;Yi Wang ,&nbsp;Chenyue Lv ,&nbsp;Hengyu Zhang ,&nbsp;Tingting Gang ,&nbsp;Chun Zhang ,&nbsp;Xiaojin Yang ,&nbsp;Baole Lu ,&nbsp;Haowei Chen","doi":"10.1016/j.yofte.2025.104291","DOIUrl":"10.1016/j.yofte.2025.104291","url":null,"abstract":"<div><div>In this paper, we report an L-band mode-locked fiber laser based on nonlinear polarization rotation (NPR). By adjusting the polarization state in the cavity, the experimentally loosely bound state (BS) can be interchanged towards the noise-like pulse (NLP). The NPR structure functions as a filter within the cavity, enabling the central wavelength of the NLP to be switched in the L-band. In addition, we numerically simulate the generation of conventional soliton, BSs, and NLP, and the results are consistent with experimental results. Our work enriches the study of the conversion of loosely BSs into NLPs based on NPR and provides new ideas for L-band light sources.</div></div>","PeriodicalId":19663,"journal":{"name":"Optical Fiber Technology","volume":"94 ","pages":"Article 104291"},"PeriodicalIF":2.6,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144195796","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An approximate policy iteration viewpoint of actor–critic algorithms","authors":"Zaiwei Chen ,&nbsp;Siva Theja Maguluri","doi":"10.1016/j.automatica.2025.112395","DOIUrl":"10.1016/j.automatica.2025.112395","url":null,"abstract":"<div><div>In this work, we establish sample complexity guarantees for a broad class of policy-space algorithms for reinforcement learning. A policy-space algorithm comprises an actor for policy improvement and a critic for policy evaluation. For the actor, we analyze update rules such as softmax, <span><math><mi>ϵ</mi></math></span>-greedy, and the celebrated natural policy gradient (NPG). Unlike traditional gradient-based analyses, we view NPG as an approximate policy iteration method. This perspective allows us to leverage the Bellman operator’s properties to show that NPG (without regularization) achieves geometric convergence to a globally optimal policy with increasing stepsizes. For the critic, we study TD-learning with linear function approximation and off-policy sampling. To address the instability of TD-learning in this setting, we propose a stable framework using multi-step returns and generalized importance sampling factors, including two specific algorithms: <span><math><mi>λ</mi></math></span>-averaged <span><math><mi>Q</mi></math></span>-trace and two-sided <span><math><mi>Q</mi></math></span>-trace. We also provide a finite-sample analysis for the critic. Combining the geometric convergence of the actor with the finite-sample results of the critic, we establish for the first time an overall sample complexity of <span><math><mrow><mover><mrow><mi>O</mi></mrow><mrow><mo>̃</mo></mrow></mover><mrow><mo>(</mo><msup><mrow><mi>ϵ</mi></mrow><mrow><mo>−</mo><mn>2</mn></mrow></msup><mo>)</mo></mrow></mrow></math></span> for finding an optimal policy (up to a function approximation error) using policy-space methods under off-policy sampling and linear function approximation.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"179 ","pages":"Article 112395"},"PeriodicalIF":4.8,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144195767","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Convolutional neural network-attention-gate recurrent unit-attention hybrid framework for spindle thermal error modeling with joint feature analysis under complex variable speed conditions","authors":"Sen Mu ,&nbsp;Guoqiang Fu ,&nbsp;Yue Zheng ,&nbsp;Xi Wang ,&nbsp;Caijiang Lu ,&nbsp;Jianzhong Fu","doi":"10.1016/j.engappai.2025.111033","DOIUrl":"10.1016/j.engappai.2025.111033","url":null,"abstract":"<div><div>Deep learning-based spindle thermal error modeling and compensation methods can effectively enhance manufacturing precision. The complex variable working conditions and thermal hysteresis effects in machine tool machining bring significant challenges for high-precision thermal error modeling. To address this issue, a hybrid structure network based on the feature extraction capability of Convolutional Neural Network (CNN) and the thermal hysteresis effect resolution capability of deep Gate Recurrent Unit (GRU) is established. A dual-layer attention mechanism is introduced to enhance spatial features and temporal features for improving the model's robustness and accuracy. First, complex variable working conditions result in complexity and nonlinearity of data. CNN is employed to extract spatial features due to its powerful feature extraction capability. A self-attention mechanism is introduced after CNN block to further filter important features. Due to the influence of thermal hysteresis effects, a deep GRU block is established to extract temporal features. A channel attention mechanism is introduced as the final layer of the network to achieve feature selection across different temperature channels. Second, features extracted by two attention mechanism layers are visualized and analyzed using t-distributed stochastic neighbor embedding (t-SNE) algorithm to explain the effectiveness of the dual-layer attention mechanism structure. The probability density distribution of the predicted results is calculated by kernel density estimation. Model performance is analyzed from the perspective of data distribution. Finally, the proposed model is compared with advanced methods under complex working conditions on two machine tools. The effectiveness of the proposed model is further validated through actual cutting compensation.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"156 ","pages":"Article 111033"},"PeriodicalIF":7.5,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144196171","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Attentive deep learning with Randomized Vector Energy Least Square Twin Support Vector Machine for Alzheimer’s Disease diagnosis","authors":"Manish Kumar ,&nbsp;Bambam Kumar ,&nbsp;Prabhat Sharma ,&nbsp;Rahul Sharma ,&nbsp;Mujahed Al-Dhaifallah ,&nbsp;Adnan Shakoor","doi":"10.1016/j.compeleceng.2025.110412","DOIUrl":"10.1016/j.compeleceng.2025.110412","url":null,"abstract":"<div><div>Alzheimer’s Disease (AD), the most common form of dementia, progressively deteriorates cognitive functions, emphasizing the importance of early and accurate diagnosis for effective treatment and management. This study proposes an advanced framework combining neuroimaging and machine learning to enhance the diagnostic precision of AD. Leveraging T1-weighted structural Magnetic Resonance Imaging (MRI) scans, the model employs a 10-layer Residual Network (ResNet) integrated with a multi-head attention mechanism to extract high-resolution features from sagittal slices, focusing on critical regions such as the hippocampus and amygdala. These features are classified using the Randomized Vector Energy Least Square Twin Support Vector Machine (RV-ELSTSVM), a novel classifier designed to improve generalization by employing randomized feature transformations and energy-based regularization. Tested on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset, the proposed framework demonstrates superior performance, achieving classification accuracies of 94.38% for CN vs AD, 88.88% for CN vs MCI, and 92.88% for MCI vs AD. By surpassing existing state-of-the-art methods, this approach highlights the efficacy of combining advanced feature extraction with robust classification techniques for early AD diagnosis. These findings pave the way for impactful clinical applications, offering healthcare professionals a powerful tool for timely intervention and management of AD. The source code of the proposed model is available at <span><span>https://github.com/rsharma2612/Randomised-SVM</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"126 ","pages":"Article 110412"},"PeriodicalIF":4.0,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144196353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Modern Academia: From “Publish or Perish” to “Monetize or Collapse”","authors":"Mohamed L. Seghier,&nbsp;Mahmoud Meribout","doi":"10.1002/ima.70137","DOIUrl":"https://doi.org/10.1002/ima.70137","url":null,"abstract":"&lt;p&gt;Academia needs money and it needs a lot and soon! Recent reports from many countries revealed that modern academia is grappling with a significant crisis in sustaining its core mission financially without burdening students with high tuition fees or relying heavily on governmental funders or private donors. This trend is more pronounced in countries like the UK than in the USA, with its strong university-industry partnerships (e.g., Silicon Valley), or in China and many European countries where universities are supported by their governments. However, with substantial cuts to government budgets for higher education, public funding is rapidly depleting, necessitating the urgent development of alternative funding models. For instance, the recent threats to some UK universities at the risk of closing whole departments and the huge loss in funding in the US to some universities and agencies [&lt;span&gt;1&lt;/span&gt;] should serve as a wake-up call for all stakeholders to prevent academia from going broke. In these uncertain times, universities are asked to make further drastic cuts or merge just to survive [&lt;span&gt;2&lt;/span&gt;].&lt;/p&gt;&lt;p&gt;The UK provides one example to gauge the true impact of financial turmoil on academia. For example, more than half of income of UK universities come from tuition fees, predominantly from international students, while a seventh of the income come from research grants (government bodies or charities) [&lt;span&gt;2&lt;/span&gt;]. According to the UK’ Office for Students, and despite an income of tens of billions of dollars, 40% of England's universities are expected to run budget deficits this year, with more than 70 universities in the UK have announced staff redundancies, department closures, programs phasing out, and other forms of restructuring [&lt;span&gt;3&lt;/span&gt;]. A similar alarming picture is also emerging in the US with research programs been closed in particular in domains judged not important by the new policy makers, as well as universities targeted with drastic cuts for not aligning with government's positions and policies [&lt;span&gt;4&lt;/span&gt;].&lt;/p&gt;&lt;p&gt;Three traditional models are gaining momentum in the current climate to save academia: expanding partnerships with industry, promoting a new breed of academic entrepreneurs, and monetising academic expertise. These models, though not new, are being administered to academia at high pace and with a degree of urgency. For instance, over the past decades, the emphasis on industrial collaborations in grant applications for science and engineering disciplines has expanded significantly, growing from a few statements to full pages. As a result, new terminology such as market maps, technological readiness levels, cost savings, competitiveness, spin-offs, patents, and marketization has become commonplace in these grant applications. Likewise, universities are establishing more incubators and frameworks to encourage their academic staff to transform their ideas and innovative solutions into marketable pro","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ima.70137","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144197402","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"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":"Multi-class classification of paint/coating defects using transfer learning","authors":"Priyankkumar Dhrangdhariya,&nbsp;Parvesh Saini,&nbsp;Soumyadipta Maiti,&nbsp;Beena Rai","doi":"10.1016/j.engappai.2025.111320","DOIUrl":"10.1016/j.engappai.2025.111320","url":null,"abstract":"<div><div>Paints and Coatings are integral to various industries, offering not only aesthetic enhancements but also protection against environmental factors like ultraviolet radiation, corrosion, and wear. However, defects compromising coating quality are common, and manual inspection methods are often inefficient, error-prone, and labour-intensive. In this study, we address these challenges by leveraging vision-based deep learning techniques for automated classification of 10 most common classes of industrial paints and coating defects. Although our dataset was limited to just 100 images (10 per class) for few-shot and 332 images (30–35 per class) for moderate-shot scenario, Vision Transformer (ViT) model achieved an impressive average accuracy of 86 % and 93 % respectively, outperforming Visual Geometry Group Network (VGG) (82 %), Densely Connected Convolutional Network (DenseNet) (91 %), and Efficient Network (EfficientNet) (79 %) in the moderate shot classification. These models’ robustness was further evaluated using a challenging non-trivial augmented test dataset designed to mimic real-world scenario. These test images were generated by mapping them onto curved surfaces, introducing distortions, and capturing from angled perspective. Remarkably, these models demonstrated robust performance with a maximum accuracy drop of only 7 %. To enhance adaptability to novel and unseen defects, Meta-Learning was also employed, improving the generalization capability of the models across both familiar and previously unseen defect types. This study offers valuable guidance for researchers working on vision-based image classification applications and provides significant insights into the potential of Artificial Intelligence (AI) driven quality control systems for the paint/coating industry, as well as in manufacturing, automotive, aerospace and civil infrastructure maintenance.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"156 ","pages":"Article 111320"},"PeriodicalIF":7.5,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144189642","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hyperparameter Control Using Fuzzy Logic: Evolving Policies for Adaptive Fuzzy Particle Swarm Optimization Algorithm.","authors":"Nicolas Roy, Charlotte Beauthier, Alexandre Mayer","doi":"10.1162/evco_a_00353","DOIUrl":"10.1162/evco_a_00353","url":null,"abstract":"<p><p>Heuristic optimization methods such as particle swarm optimization (PSO) depend on their parameters to achieve optimal performance on a given class of problems. Some modifications of heuristic algorithms aim at adapting those parameters during the optimization process. We present a novel approach to design such adaptation strategies using continuous fuzzy feedback control. Fuzzy feedback provides a simple interface where probes are sampled in the optimization process and parameters are fed back to the optimizer. The probes are turned into parameters by a fuzzy process optimized beforehand to maximize performance on a training benchmark. Utilizing this framework, we systematically established 127 different fuzzy PSO algorithms featuring a maximum of seven parameters under fuzzy control. These newly devised algorithms exhibit superior performance compared to both traditional PSO and some of its best parameter control variants. The performance is reported in the single-objective bound-constrained numerical optimization competition of Congress on Evolutionary Computation (CEC) 2020. Additionally, two specific controls, highlighted for their efficacy and dependability, demonstrated commendable performance in real-world scenarios from CEC 2011.</p>","PeriodicalId":50470,"journal":{"name":"Evolutionary Computation","volume":" ","pages":"279-308"},"PeriodicalIF":4.6,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141421719","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Leak Detection of Underground Water Pipelines Using Acoustic Feature Extraction","authors":"Shunyi Zhao, Qingxin Lu, Tianyu Zhang, Shuping He, Peng Shi, Jionghui Li","doi":"10.1109/jiot.2025.3575465","DOIUrl":"https://doi.org/10.1109/jiot.2025.3575465","url":null,"abstract":"","PeriodicalId":54347,"journal":{"name":"IEEE Internet of Things Journal","volume":"29 1","pages":""},"PeriodicalIF":10.6,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144201663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}