检索结果-内蒙古大学图书馆

15th IEEE International Conference on Wireless Communications and Signal Processing, WCSP 2023

作者： Wang, Siyuan Yan, Qifa Zhang, Jingjing Wang, Jianping Song, Linqi City University of Hong Kong Department of Computer Science Hong Kong Southwest Jiaotong University The School of Information Science and Technology Chengdu China Fudan University Department of Communication Science and Engineering Shanghai China

ISBN: (纸本)9798350324662

Low-rank matrix factorization (MF) decomposes a matrix into two low-dimensional matrices. One key algorithm for MF is distributed alternating least squares (ALS), iteratively learning the matrices. However, hardware, software, or network problems can cause distributed nodes (stragglers) to fail computation or communication. Existing algorithms neglect stragglers as a design concern. To address this, we propose D-ALS-MC (Distributed ALS Matrix Completion), a coding-assisted algorithm for distributed matrix completion. We analyze the trade-off between the maximum tolerable number of stragglers (recovery threshold) and coding redundancy (code rate), along with computational costs. Moreover, we prove that incomplete matrices satisfying the Restricted Isometry Property (R.I.P.) condition can recover original values. The experimental results demonstrate the effectiveness of our proposed methods. © 2023 IEEE.

关键词： coded computation federated learning matrix completion

来源：评论

学校读者我要写书评

暂无评论

Detection of anomalies in networks by implementing and analyzing deep learning-based classifiers

Detection of anomalies in networks by implementing and analy...

引用

2024 International Conference on Communication, Computing and Internet of Things, IC3IoT 2024

作者： Anitha, Cuddapah Aswal, Upendra Singh Chaithra, M.H. Dange, Jyoti Barve, Amit Maranan, Ramya Mohan Babu University Erstwhile Sree Vidyanikethan Engineering College School Of Computing Department Of Computer Science And Engineering Andhra Pradesh Tirupati517102 India Graphic Era Deemed To Be University Department Of Computer Science & Engineering Uttarakhand Dehradun248002 India Reva University Department Of Computer Science And Engineering Karnataka Bangalore560064 India Atharva College Of Engineering Department Of Information Technology Malad West Maharashtra Mumbai400095 India Parul University Parul Institute Of Technology Faculty Of Engineering And Technology Department Of Computer Science And Engineering P.O.Limda Gujarat 391760 India Saveetha School Of Engineering Simats Department Of Research And Innovation Tamil Nadu Chennai602105 India

ISBN: (纸本)9798350352689

With the increasing complexity and sophistication of cyber threats, the demand for effective anomaly detection mechanisms in network security has grown substantially. This research focuses on the utilization of deep learning (DL) to detect anomalies in networks, implementing and analyzing three distinct models of deep learning-based classifiers. The classifiers include a binary classification model, a multiclass classification model with proportionality classification, and a multiclass classification model with exclusion strategies. The efficacy of these models is evaluated through cross-validation runs and testing on data from a database parallel to that of its training. The binary classification model serves as the baseline, distinguishing between normal and anomalous network behavior. The multiclass classification model with proportionality classification extends the binary approach by categorizing anomalies into multiple classes with consideration for their proportional representation. On the other hand, the multiclass classification model with exclusion strategies refines anomaly detection by explicitly excluding certain patterns from classification, enhancing the model's specificity. The research methodology involves the development of deep learning-based classifiers, the collection of network data for training, and rigorous evaluation through cross-validation techniques. Additionally, the models are tested on a separate dataset to assess their generalization capabilities in real-world scenarios. The outcomes of this research contribute to the understanding of DL applications in network anomaly detection, providing insights into the strengths and limitations of different classifier strategies. © 2024 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Remote Sensing Image Object Detection by Fusing Multi-Scale Contextual Features and Channel Enhancement

Remote Sensing Image Object Detection by Fusing Multi-Scale ...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Xuesen Ma Jindian Dong Weixin Wei Biao Zheng Ji Ma Tianbao Zhou School of Computer Science and Information Engineering Hefei University of Technology Hefei China Intelligent Interconnected Systems Laboratory of Anhui Province Hefei University of Technology Hefei China

Remote sensing technology is becoming more sophisticated and is extensively used for object tracking, urban planning, military reconnaissance and other fields. Complex backgrounds and diverse object scales are two important factors that affect the object detection effect of remote sensing images. To address this problem, this paper proposes a remote sensing object detection model that incorporates channel enhancement and multi-scale contextual features. Firstly, the multi-scale contextual feature enhancement module is constructed, which performs multi-order spatial interaction by cascading recursive convolution to obtain contextual information of feature maps at different scales, and introduces attention to reinforce unique features of objects and suppress background interference. Then, the spatial pyramid channel enhancement module combining sub-pixel convolution and adaptive sampling factor is designed to mitigate the semantic weakening of the depth feature maps caused by channel downscaling, thus enhancing the sampling effect between feature maps of different scales and reducing information loss. Finally, the effectiveness of the model is verified on the large-scale remote sensing image object detection dataset DIOR.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Lightweight Convolutional Network For Automated Photovoltaic Defect Detection 9

Lightweight Convolutional Network For Automated Photovoltaic...

引用

9th International Conference on information technology Trends, ITT 2023

作者： Zahid, Arsalan Hussain, Muhammad Hill, Richard Al-Aqrabi, Hussain University of Huddersfield School of Computing and Engineering Huddersfield United Kingdom Higher Colleges of Technology Faculty of Computer Information Science Sharjah United Arab Emirates

ISBN: (纸本)9798350327502

As the World moves towards renewable energy, photovoltaic modules are a fundamental option due to their green nature. However, the manufacturing process of solar cells is complex and vulnerable to discrepancies which can impact the overall performance of the system. Although human-led inspection is seen as the de-facto quality inspection protocol, issues pertaining to bias, cost and time can make it an expensive process. To this effect, this paper focuses on the development of a custom convolutional architecture that is lightweight, hence deployable within manufacturing facilities to assist with defective solar cell inspection. In addition, to address the issue of data scarcity, representative data augmentations are producing tailored towards enhancing the model's generalizability. The high efficacy of the proposed CNN and proposed augmentations can be gauged by the fact that 98% F1 score was achieved overall. © 2023 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Solar Observation with the Fourier Transform *** Results of Solar Spectrum near the CO 4.66μm and Mg I 12.32μm

引用

Research in Astronomy and Astrophysics 2023年第12期23卷 128-137页

作者： Xianyong Bai Zhiwei Feng Zhiyong Zhang Yuanyong Deng Xiaoming Zhu Kaifan Ji Song Feng Xiao Yang Yongliang Song National Astronomical Observatories Chinese Academy of SciencesBeijing 100101China School of Astronomy and Space Science University of Chinese Academy of SciencesBeijing 101408China Yunnan Observatories Chinese Academy of SciencesKunming 650011China Faculty of Information Engineering and Automation/Yunnan Key Laboratory of Computer Technology Application Kunming University of Science and TechnologyKunming 650500China

An infrared solar spectrum observed by ground-based telescopes is seriously affected by the background radiation both from the telescope and sky,relative to the visible *** accuracy is also influenced by the spectral resolution of the Fourier transform *** the paper,we developed a CO_(2)gas cell and installed it in the sample compartment to calibrate the spectral resolution of the Bruker IFS-125HR at infrared *** measured spectral resolution is 0.00342±0.00086 cm^(-1)and 0.0059±0.00024 cm^(-1)at the wavenumbers of798 cm^(-1)and 2136 cm^(-1),*** also updated a fully reflective sunlight feeding system to observe the solar spectrum near CO 4.66μm and Mg I 12.32μ*** quickly pointing the sunlight feeding system about1 degree away from the solar disk center,we are able to measure the background radiation from the telescope and the sky at Huairou Solar Observing *** removing the background radiation,our observed solar spectrum at CO 4.66μm is consistent with that from the National Solar *** Mg I 12.32μm working line selected by the Accurate Infrared Magnetic Field Measurements of the Sun(AIMS)project is also *** method is helpful not only for the spectral resolution calibration and background radiation correction of AIMS but also for other infrared astronomical telescopes.

关键词： Sun infrared-methods observational-instrumentation interferometers

来源：评论

学校读者我要写书评

暂无评论

Three-stage data generation algorithm for multiclass network intrusion detection with highly imbalanced dataset

引用

International Journal of Intelligent Networks 2023年第1期4卷 202-210页

作者： Chui, Kwok Tai Gupta, Brij B. Chaurasia, Priyanka Arya, Varsha Almomani, Ammar Alhalabi, Wadee Department of Electronic Engineering and Computer Science School of Science and Technology Hong Kong Metropolitan University Hong Kong Taichung 413 Taiwan Department of Computer Science and Information Engineering Asia University Taichung 413 Taiwan School of Computing Engineering and Intelligent Systems Ulster University United Kingdom Department of Business Administration Asia University Taiwan School of Information Technology Skyline University College SharjahP.O. Box 1797 United Arab Emirates Al- Balqa Applied University Jordan Immersive Virtual Reality Research Group King Abdulaziz University Jeddah21589 Saudi Arabia Symbiosis International University Pune India Lebanese American University Beirut1102 Lebanon Birkbeck University of London United Kingdom Chandigarh University Chandigarh India

The Internet plays a crucial role in our daily routines. Ensuring cybersecurity to Internet users will provide a safe online environment. Automatic network intrusion detection (NID) using machine learning algorithms has recently received increased attention recently. The NID model is prone to bias towards the classes with more training samples due to highly imbalanced datasets across different types of attacks. The challenge in generating additional training data for minority classes is the generation of insufficient data. The study's purpose is to address this challenge, which extends the data generation ability by proposing a three-stage data generation algorithm using the synthetic minority over-sampling technique, a generative adversarial network (GAN), and a variational autoencoder. A convolutional neural network is employed to extract the representative features from the data, which were fed into a support vector machine with a customised kernel function. An ablation study evaluated the effectiveness of the three-stage data generation, feature extraction, and customised kernel. This was followed by a performance comparison between our study and existing studies. The findings revealed that the proposed NID model achieved an accuracy of 91.9%–96.2% in the four benchmark datasets. In addition, it outperformed existing methods such as GAN-based deep neural networks, conditional Wasserstein GAN-based stacked autoencoder, synthesised minority oversampling technique-based random forest, and variational autoencoder-based deep neural network, by 1.51%–28.4%. © 2023 The Authors

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Text-guided Reconstruction Network for Sentiment Analysis with Uncertain Missing Modalities

引用

IEEE Transactions on Affective Computing 2025年

作者： Shi, Piao Hu, Min Nakagawa, Satoshi Zheng, Xiangming Shi, Xuefeng Ren, Fuji Hefei University of Technology Key Laboratory of Knowledge Engineering with Big Data Ministry of Education Anhui Province Key Laboratory of Affective Computing and Advanced Intelligent Machine National Smart Eldercare International Science and Technology Cooperation Base School of Computer Science and Information Engineering Anhui Hefei230601 China Bozhou University School of Electronic and Information Engineering Bozhou236800 China University of Tokyo Graduate School of Information Science and Technology Tokyo113-8656 Japan University of Electronic Science and Technology of China College of Computer Science and Engineering Chengdu611731 China University of Electronic Science and Technology of China Shenzhen Institute for Advanced Study Shenzhen518110 China

Multimodal Sentiment Analysis (MSA) is an attractive research that aims to integrate sentiment expressed in textual, visual, and acoustic signals. There are two main problems in the existing methods: 1) the dominant role of the text is underutilization in unaligned multimodal data, and 2) the modality under uncertain missing feature is not sufficiently explored. This paper proposes a Text-guided Reconstruction Network (TgRN) for MSA with uncertain missing modalities in non-aligned sequences. The TgRN network includes three primary modules: Text-guided Extraction Module (TEM), Reconstruction Module (RM) and Text-guided Fusion Module (TFM). First, the TEM consists of the text-guided cross attention units and self-attention units to capture inter-modal features and intra-modal features, respectively. Second, leveraging enhanced attention units and a three-way squeeze-and-excitation block, the RM is designed to learn semantic information from incomplete data and reconstruct missing modality features. Third, the TFM utilizes a progressive modality-mixing adaptation gate to explore the dynamic correlations between nonverbal and verbal modalities, effectively addressing the modality gap issue. Finally, under the supervision of sentiment prediction loss and reconstruction loss, the TgRN effectively processes both uncertain missing-modality conditions and ideal complete modality conditions. Extensive experiments on CMU-MOSI and CH-SIMS demonstrate that our proposed method outperforms state-of-the-art approaches. © 2010-2012 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

NOFTrack: Efficient Multi-Object Tracking with a Speed Advantage

NOFTrack: Efficient Multi-Object Tracking with a Speed Advan...

引用

2024 International Symposium on Digital Home, ISDH 2024

作者： Jiang, Ming Ke, Jiehao Guo, Biao Metaverse Application Engineering Center Guilin University of Electronic Technology Guilin China School of Computer Science and Information Security Guilin University of Electronic Technology Guilin China Guilin University of Electronic Technology Guangxi Key Laboratory of Trusted Software Guilin China

ISBN: (纸本)9798331509873

In the field of multi-object tracking (MOT), accurately tracking the position and characteristics of targets is crucial. However, traditional methods often overlook detection results with low confidence, which can lead to the omission of targets and the fragmentation of trajectories. To overcome these challenges, we have introduced a novel NOF algorithm. This algorithm efficiently handles detection boxes through a tiered strategy, initially using an adaptive Kalman filter for preliminary matching, followed by a secondary screening of detection boxes with lower confidence, and a final stage for processing dubious detection results. This approach not only enhances the speed and accuracy of tracking but also performs exceptionally well when dealing with complex scenes. Moreover, it is capable of achieving high efficiency, with a rate of over 300 frames per second (FPS), ensuring the efficiency and precision of the tracking process. © 2024 IEEE.

关键词： Target tracking

来源：评论

学校读者我要写书评

暂无评论

GraphUnet: Graph Make Strong Encoders for Remote Sensing Segmentation 29

GraphUnet: Graph Make Strong Encoders for Remote Sensing Seg...

引用

29th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2023

作者： Shou, Yuntao Ai, Wei Meng, Tao Zhang, Fuchen Li, Keqin Central South University of Forestry and Technology School of Computer and Information Engineering ChangSha China State University of New York Department of Computer Science New PaltzNY12561 United States

ISBN: (纸本)9798350330717

Remote sensing segmentation are widely applied in environmental protection, and urban change detection, etc. Despite the success of deep learning-based remote sensing segmentation methods (e.g., CNN and Transformer), they are not flexible enough to model irregular objects. To tackle the above problems, this paper treats images as graph structures and introduces a simple vision GNN (GraphUNet) architecture for remote sensing segmentation. Specifically, we construct a graph view and utilize GNN to extract the visual feature. To extract the global contextual location information in the image, we introduce a multi-head attention mechanism for global information extraction. Furthermore, we introduce a feed-forward network for each node to perform feature transformation on node features to encourage information diversity. Extensive experiments on three real datasets demonstrate that our model GraphUNet outperforms SOTA remote sensing image segmentation methods. © 2023 IEEE.

关键词： Graph Learning Multi-head attention Remote Sensing Segentation U-shaped structure

来源：评论

学校读者我要写书评

暂无评论

An Ensemble Fake News Detection Model Using fakenewslingual dataset with Feature Selection 12

An Ensemble Fake News Detection Model Using fakenewslingual ...

引用

12th IEEE Conference on Systems, Process and Control, ICSPC 2024

作者： Choong, Hoi Chun Chua, Hui Na Jasser, Muhammed Basheer Wong, Richard T.K. Aldharhani, Ghassan Saleh Sunway University No. 5 School of Engineering and Technology Department of Computing and Information Systems Jalan Universiti Selangor Darul Ehsan Bandar Sunway47500 Malaysia Institute of Computer Science and Digital Innovation UCSI University Kuala Lumpur Malaysia UKM University Faculty of Information Science and Technology Selangor Bangi Malaysia

ISBN: (纸本)9798350391398

The rapid dissemination of information on social media acts as a double-edged sword, promoting the spread of accurate information and fake news, which can disrupt social cohesion and undermine trust in media and institutions. Existing fake news detection methods often lack feature comprehensiveness and struggle to generalize across different fake news types and sources. This study aims to enhance feature comprehensiveness by incorporating headline features into existing models and improve model generalization through ensemble techniques and hyperparameter tuning. The study expanded an existing dataset with headline, sentiment, and readability features. We assessed the impact of these additional features on model performance and explored various ensemble learning techniques (Bagging, AdaBoost, Gradient Boost) with tailored hyperparameters. An optimal model was then used to build a prototype for real and fake news classification. Integrating headline features showed marginally improved accuracy from 98.63% to 98.66%. While using ensemble learning models and hyperparameter tuning, accuracy increased significantly to 99.34%. This research enriched the dataset with headline features, enhancing the detection model's accuracy and generalization. The enhanced dataset, which we named fakenewslingual, has been made publicly available for further research. The study underscores the effectiveness of headline features and ensemble models in fake news detection. © 2024 IEEE.

关键词： Data accuracy

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：