检索结果-内蒙古大学图书馆

Dependency-augmented graph aggregation networks for document-level relation extraction with cross-sentence semantics

引用

Journal of King Saud University - Computer and Information Sciences 2025年第3期37卷 1-12页

作者： Che, Shasha He, Qing Yang, Zhihao Li, Yanbo Du, Nisuo College of Big Data and Information Engineering Guizhou University Guiyang China Guizhou Provincial Key Laboratory of Public Big Data Guizhou University Guiyang China

Finding semantic relationships between words in several sentences is the goal of document-level relation extraction (DocRE), a crucial problem in natural language processing. Current research is unable to accurately characterize long-range interdependence and cross-sentence interactions, which restricts their capacity to capture document-level semantics. In order to alleviate this issue, we introduce the Dependency-Augmented Graph Aggregation Network(DAGA), which is a novel DocRE model. In particular, we design a Dependency Graph Aggregation Module (DGAM) that integrates Sentence Related Graph and Dependency Structure Graph to explore both local and global relational patterns. To explicitly capture document-level sentence-related dependencies and semantic interactions, we propose Dependency-Augmented Attention Mechanism (DAAM). Results from experiments show that our suggested approach improves the F1 score by 1.39 and the Ign F1 score by 1.52 on publicly available benchmark datasets. In summary, DAGA demonstrates higher performance in dealing with complicated semantic relationships at the document level. © The Author(s) 2025.

关键词： Cross-sentence relationships Dependency structure graph Document-level relation extraction Multi-head attention Semantic relations

来源：评论

学校读者我要写书评

暂无评论

FedDAGC: Dynamic Adaptive Graph Coarsening for Federated Learning on Non-IID Graphs 18th

FedDAGC: Dynamic Adaptive Graph Coarsening for Federated Le...

引用

18th International Conference on Wireless Artificial Intelligent Computing Systems and Applications, WASA 2024

作者： Zhang, Chengxi Hu, Chunqiang College of Big Data and Software Engineering Chongqing University Chongqing China

ISBN: (纸本)9783031714665

Federated graph learning (FGL) has demonstrated great potential in collaboratively training graph neural network (GNN) models under the federated learning (FL) framework, benefiting from its advantage of privacy preservation. However, FGL suffers from the significant challenge of non-IID problem. In the real world, cross-source or even cross-domain graph data frequently exist among clients. To tackle this challenge, we propose a novel FGL framework, FedDAGC, from the new perspective of graph coarsening and consensus. Specifically, our method inserts a dynamic adaptive graph coarsening (DAGC) module into local GNN models to generate coarsened graphs as augmented inputs for subsequent model training. By sharing the intermediate parameters in DAGC instead of traditional GNN model parameters, FedDAGC enables clients to reach a consensus on coarsening directions, for extracting crucial patterns from local graphs based on common core knowledge while protecting privacy. Extensive experiments and analyses on various non-IID settings demonstrate the efficacy and superiority of FedDAGC. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Clustering ensemble pruning algorithm based on vec2vec classifier representation

引用

Journal of King Saud University - Computer and Information Sciences 2025年第3期37卷 1-21页

作者： Li, Chunlong Li, Danyang Qin, Xue Qin, Yuanxian Li, Jialin Yuan, Jiafan College of Big Data and Information Engineering Guizhou University Jiaxiu South Road Guiyang 550025 China

The complementarity of ensemble learning primarily stems from the diversity among its classifier members, which is typically assessed by calculating the similarity or distance between classifiers. However, traditional methods for similarity or distance calculation are limited by the representation of classifiers, which may not accurately reflect the true relationships between them, leading to inferior ensemble pruning performance. To address this issue, this study proposes a new classifier representation paradigm and ensemble pruning algorithm(Vec2Q-CEEP). To begin with, we transform the representation of classifiers through spatial mapping, overcoming the limitations of traditional similarity calculation methods, enriching the ways of calculating classifier diversity, and enhancing the distinguishability between classifiers. Furthermore, we introduce a novel similarity measurement method that integrates classifier information from both the original and mapped spaces, enabling a more comprehensive evaluation of the relationships between classifiers. After conducting extensive experiments on 12 UCI datasets, the experimental results demonstrate that this algorithm outperforms several state-of-the-art ensemble pruning algorithms, thus validating its effectiveness and feasibility. © The Author(s) 2025.

关键词： Classifier ensemble Classifier representation Classifier selection Ensemble learning Ensemble pruning

来源：评论

学校读者我要写书评

暂无评论

Steel Surface Defect Detection Using Learnable Memory Vision Transformer

引用

Computers, Materials & Continua 2025年第1期82卷 499-520页

作者： Syed Tasnimul Karim Ayon Farhan Md.Siraj Jia Uddin Department of Computer Science and Engineering BRAC UniversityDhaka1212Bangladesh Department of AI and Big Data Endicott CollegeWoosong UniversityDaejeon34606Republic of Korea

This study investigates the application of Learnable Memory Vision Transformers(LMViT)for detecting metal surface flaws,comparing their performance with traditional CNNs,specifically ResNet18 and ResNet50,as well as other transformer-based models including Token to Token ViT,ViT withoutmemory,and Parallel *** awidely-used steel surface defect dataset,the research applies data augmentation and t-distributed stochastic neighbor embedding(t-SNE)to enhance feature extraction and *** techniques mitigated overfitting,stabilized training,and improved generalization *** LMViT model achieved a test accuracy of 97.22%,significantly outperforming ResNet18(88.89%)and ResNet50(88.90%),aswell as the Token to TokenViT(88.46%),ViT without memory(87.18),and Parallel ViT(91.03%).Furthermore,LMViT exhibited superior training and validation performance,attaining a validation accuracy of 98.2%compared to 91.0%for ResNet 18,96.0%for ResNet50,and 89.12%,87.51%,and 91.21%for Token to Token ViT,ViT without memory,and Parallel ViT,*** findings highlight the LMViT’s ability to capture long-range dependencies in images,an areawhere CNNs struggle due to their reliance on local receptive fields and hierarchical feature *** additional transformer-based models also demonstrate improved performance in capturing complex features over CNNs,with LMViT excelling particularly at detecting subtle and complex defects,which is critical for maintaining product quality and operational efficiency in industrial *** instance,the LMViT model successfully identified fine scratches and minor surface irregularities that CNNs often *** study not only demonstrates LMViT’s potential for real-world defect detection but also underscores the promise of other transformer-based architectures like Token to Token ViT,ViT without memory,and Parallel ViT in industrial scenarios where complex spatial relationships are *** research m

关键词： Learnable Memory Vision Transformer(LMViT) Convolutional Neural Networks(CNN) metal surface defect detection deep learning,computer vision image classification learnable memory gradient clipping label smoothing t-SNE visualization

来源：评论

学校读者我要写书评

暂无评论

Optimized Wavelet Neural Networks Based on Tunicate Swarm Algorithm for Short-Term Traffic Flow Prediction

Journal of Network Intelligence

引用

Journal of Network Intelligence 2025年第1期10卷 299-311页

作者： Li, Jian-Fan Li, Zhong-Wu Fu, Arends College of Transportation Engineering Tongji University Shanghai201804 China College of Big Data Baoshan University Baoshan678000 China College of Engineering and Information Technology Cavite State University Lahug Cebu6000 Philippines

Short-term traffic flow prediction is an important research area in smart transportation systems. To enhance prediction accuracy, scholars have been exploring different combination models. Among these models, the wavelet neural network stands out as it combines wavelet analysis and neural networks, offering improved results in short-term traffic flow forecasting. The tunicate group optimization algorithm is another promising technique, characterized by its simplicity, efficiency, and minimal parameters required. In this study, we introduce the tunicate swarm algorithm to adjust the weights and wavelet factors of the network in the Wavelet Neural Networks (WNN) model, aiming to enhance its prediction accuracy and overall performance. We first apply wavelet analysis to distill the key features of the traffic flow data. Next, we construct the WNN model and input the traffic flow data, along with data from previous moments, for training and learning. Subsequently, we optimize the network’s weights and wavelet factors using the Tunicate Swarm Algorithm (TSA), further enhancing the model’s performance. Experimental findings showcase that our suggested technique attains remarkable precision and consistency in short-term traffic flow forecasting. Thus, this method holds significant value in traffic management and planning, providing practical applications for improving the reliability and accuracy of traffic flow forecasting. © 2025, Taiwan Ubiquitous Information CO LTD. All rights reserved.

关键词： Short traffic flow prediction Tunicate swarm algorithm Wavelet neural network

来源：评论

学校读者我要写书评

暂无评论

YOLOv9-AAG: Distinguishing Birds and Drones in Infrared and Visible Light Scenarios

引用

IEEE Access 2025年 13卷 76609-76619页

作者： Suo, Peichun Zhu, Jincan Zhou, Qingyang Kou, Weili Wang, Xiuyan Suo, Wen Yunnan Medical Health Vocational College Information Center Kunming650033 China Southwest Forestry University College of Big Data and Intelligent Engineering Kunming650224 China Key Laboratory of State Forestry and Grassland Administration on Forestry Ecological Big Data Kunming650224 China Yunnan College of Business Management Kunming650106 China

The ability to distinguish between birds and drones is essential for applications in wildlife preservation, aviation security, and defense operations. Reliable identification not only reduces the risk of bird strikes and monitors potential drone threats but also fosters the development of a secure and intelligent ecological environment. However, achieving high precision in recognition under complex and dynamic conditions remains a significant hurdle. To overcome this, we present an advanced detection framework, YOLOv9-AKConv-AFF-GSConv(YOLOv9-AAG), which substantially improves the accuracy of distinguishing birds from drones, even in varied poses, and enhances generalization capabilities in both infrared and visible light scenarios. The framework incorporates several key advancements. First, the GSConv is embedded into the head of the RepNSCPELAN4-AKConv module, enabling efficient extraction of edge and contour information without introducing additional computational cost. Second, the inclusion of the Attentional Feature Fusion (AFF) mechanism within AKConv amplifies the representation of shape features, particularly under challenging environments and multimodal scenarios. Lastly, the optimized AKConv module is effectively integrated with the original RepNSCPELAN4 model, facilitating more accurate and robust feature extraction for small targets, including birds and drones. Experimental evaluations underscore the effectiveness of the proposed framework. On an optical dataset containing multiple bird and drone targets, YOLOv9-AAG achieved recognition accuracy rates of 78.2%, while on an infrared dataset, it reached 86.1%. These results reflect improvements of 7.4% and 6.0%, respectively, over the baseline YOLOv9 model. Additionally, the F1 scores increased by 5.9% and 2.8%. On the publicly available CUB-200-2011 dataset and a combined dataset of visible and infrared images, the model attained mean Average Precision (mAP) scores of 81.9% and 81.6%, respectively, outper

关键词： Birds

来源：评论

学校读者我要写书评

暂无评论

Reconfigurable metasurface for code division multiplexing in high-capacity and low-crosstalk holographic encryption

引用

Optics Express 2025年第4期33卷 7295-7306页

作者： Li, Rui Yuan, Huan Zhong, Zheqiang Zhang, Bin College of Electronics and Information Engineering Sichuan University Sichuan Chengdu610064 China Big Data Analysis and Fusion Application Technology Engineering Laboratory of Sichuan Province Chengdu610065 China

Metasurface multiplexing holography shows potential to enhance information capacity but faces challenges such as crosstalk and limited dynamic switching. Leveraging the orthogonality of code division multiplexing (CDM) technology, we propose a reconfigurable metasurface-based encryption scheme. Here, the CDM code acts as a decryption key, enabling dynamic holographic image switching and ensuring secure information transmission. The circular polarization information is encoded into the reconfigurable metasurface using simultaneous modulation of geometric and propagation phases. By controlling polarization states of incident light and crystallization levels of GSST, the high-capacity and low-crosstalk CDM holographic images can be generated. Furthermore, a detailed analysis is conducted on the capacity of CDM holography, the factors influencing the efficiency of metasurface diffraction, and the scalability and feasibility of practical manufacturing. This proposed scheme has the ability to significantly boost encryption security, paving the way for increased information storage capacity and optical encryption. © 2025 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement.

关键词： Crosstalk

来源：评论

学校读者我要写书评

暂无评论

CASSNet: Cross-Attention Enhanced Spectral-Spatial Interaction Network for Hyperspectral Image Super-Resolution

引用

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2025年 18卷 11716-11730页

作者： Zhang, Zhanxu Yang, Linzi Zhang, Guanlian Deng, Jiangwei Bian, Lifeng Yang, Chen Guizhou University Power Systems Engineering Research Center Ministry of Education College of Big Data and Information Engineering Guiyang550025 China Guizhou University China State Key Laboratory of Public Big Data Guiyang550025 China

Deep learning-based super-resolution (SR) methods for single hyperspectral image have made significant progress in recent years and become an important research direction in remote sensing. Existing methods perform well in extracting spatial features, but challenges remain in integrating spectral and spatial features when modeling global relationships. In order to take full advantage of the higher spectral resolution of hyperspectral images, this paper proposes a novel hyperspectral image SR method (CASSNet), which integrates convolutional neural networks and cross-attention mechanisms into a unified framework. This approach achieves comprehensive integration of spectral and spatial information, with extensive exploration at both local and global levels. In the local feature extraction stage, parallel 3D/2D convolutions work in tandem to efficiently capture detail information from both spectral and spatial dimensions. In addition, a spectral-spatial dual-branch module employing the cross-attention mechanism is designed to capture the global dependencies within the features. Where the reconstructed spectral-spatial module and the spectral-spatial interaction unit can effectively promote the interaction and complementarity of spectral-spatial features. The experiments on three publicly available datasets demonstrated that the proposed method obtained superior SR results, outperforming state-of-the-art SR algorithms. © 2008-2012 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A survey on cross-user federated recommendation

引用

Science China(Information Sciences) 2025年第4期68卷 7-32页

作者： Enyue YANG Yudi XIONG Wei YUAN Weike PAN Qiang YANG Zhong MING College of Computer Science and Software Engineering Shenzhen University School of Electrical Engineering and Computer Science The University of Queensland WeBank AI Lab WeBank Department of Computer Science and Engineering Hong Kong University of Science and Technology College of Big Data and Internet Shenzhen Technology University Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ)

Recommender systems are effective in mitigating information overload, yet the centralized storage of user data raises significant privacy concerns. Cross-user federated recommendation(CUFR) provides a promising distributed paradigm to address these concerns by enabling privacy-preserving recommendations directly on user devices. In this survey, we review and categorize current progress in CUFR, focusing on four key aspects: privacy, security, accuracy, and efficiency. Firstly,we conduct an in-depth privacy analysis, discuss various cases of privacy leakage, and then review recent methods for privacy protection. Secondly, we analyze security concerns and review recent methods for untargeted and targeted *** untargeted attack methods, we categorize them into data poisoning attack methods and parameter poisoning attack methods. For targeted attack methods, we categorize them into user-based methods and item-based methods. Thirdly,we provide an overview of the federated variants of some representative methods, and then review the recent methods for improving accuracy from two categories: data heterogeneity and high-order information. Fourthly, we review recent methods for improving training efficiency from two categories: client sampling and model compression. Finally, we conclude this survey and explore some potential future research topics in CUFR.

关键词： cross-user federated recommendation federated recommendation federated learning recommender systems user privacy

来源：评论

学校读者我要写书评

暂无评论

kProtoClust:Towards Adaptive k-Prototype Clustering without Known k

引用

Computers, Materials & Continua 2025年第3期82卷 4949-4976页

作者： Yuan Ping Huina Li Chun Guo Bin Hao School of Information Engineering Xuchang UniversityXuchang461000China Henan Province Engineering Technology Research Center of Big Data Security andApplications Xuchang461000China College of Computer Science and Technology Guizhou UniversityGuiyang550025China Here Data Technology Shenzhen518000China

Towards optimal k-prototype discovery,k-means-like algorithms give us inspirations of central samples collection,yet the unstable seed samples selection,the hypothesis of a circle-like pattern,and the unknown K are still challenges,particularly for non-predetermined data *** propose an adaptive k-prototype clustering method(kProtoClust)which launches cluster exploration with a sketchy division of K clusters and finds evidence for splitting and *** behalf of a group of data samples,support vectors and outliers from the perspective of support vector data description are not the appropriate candidates for prototypes,while inner samples become the first candidates for instability reduction of *** from the representation of samples in traditional,we extend sample selection by encouraging fictitious samples to emphasize the representativeness of *** get out of the circle-like pattern limitation,we introduce a convex decomposition-based strategy of one-cluster-multiple-prototypes in which convex hulls of varying sizes are prototypes,and accurate connection analysis makes the support of arbitrary cluster shapes *** by geometry,the three presented strategies make kProtoClust bypassing the K dependence well with the global and local position relationship analysis for data *** results on twelve datasets of irregular cluster shape or high dimension suggest that kProtoClust handles arbitrary cluster shapes with prominent accuracy even without the prior knowledge K.

关键词： Prototype finding convex hull support vector data description geometrical information

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：