检索结果-内蒙古大学图书馆

21st International Learning and technology Conference, L and T 2024

作者： Preethi, I. Dharmarajan, K. Sharma, Bhisham Chowdhury, Subrata Dhaou, Imed Ben Vistas School of Computing Sciences Chennai India Vistas Department of Information Technology Chennai India Chitkara University Institute of Engineering and Technology Chitkara University Department of Computer Science and Engineering Punjab India Sreenivasa Institute of Technology and Management Studies Department of Computer Science and Engineering Andhra Pradesh Chittoor India Dar Al-Hekma University Hekma School of Engineering Computing Design Department of Computer Science Jeddah Saudi Arabia

ISBN: (纸本)9798350393569

Chronic Kidney Disease (CKD), a significant global health issue and has a major impact on a vast number of people. In the early stages, it is impossible to identify the disease's symptoms. Very few people are aware of this illness and are able to foresee the symptoms earlier. The disorder is considered as potential risk factor. As a result, a potent deep learning framework is used in this paper to reduce the risk of acquiring these diseases as a way to prevent them. The machine learning algorithms used in early disease prediction, has been found to be computationally expensive, frequently overfit, and underperform in terms of accuracy since they must examine the large amount of clinical data until the model converges. Therefore, in this paper three novel work has been proposed an efficient novel hybrid feature selection strategy RFITLO algorithm is used to find the optimal features that gives the major contribution is classifying the CKD disease. Then two proposed classification algorithms namely Enhanced Multi-Layer Perceptron (HW-MLP) and Optimized Multi-Layer Perceptron (PKD-OMLP) are used in prediction model to capture the complex patterns and optimize the learning algorithm to predict the CKD at prior stage from the data gathered in Kaggle, Real Time and UCI Machine Learning Repository Dataset. In order to measure the classifications of disease, performance measures including accuracy, precision and recall are analyzed. The experimental findings show that the PKD-OMLP strategy produces better results than the proposed HW-MLP and other conventional approaches like Support Vector Machine (SVM), Linear Regression (LR) and Multi-Layer Perceptron (MLP). Among the preceding four models PKD-OMLP renders the best outcome as per its performance level producing a high accuracy of 94.89% on testing Real Time (RT) CKD dataset comparatively with other datasets such as Kaggle and UCI repository. Therefore, these proposed algorithms can support clinicians to enable secured an

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Resource-Limited Localized Adaptive Adversarial Training for Machine Learning Model 21

Resource-Limited Localized Adaptive Adversarial Training for...

引用

21st IEEE International Symposium on Parallel and Distributed Processing with Applications, 13th IEEE International Conference on Big Data and Cloud computing, 16th IEEE International Conference on Social computing and Networking and 13th International Conference on Sustainable computing and Communications, ISPA/BDCloud/SocialCom/SustainCom 2023

作者： Rajhi, Mohammed Pissinou, Niki Florida International University School of Computing and Information Sciences MiamiFL United States Jazan University Faculty of Computer Science & Information Technology Jazan Saudi Arabia

ISBN: (纸本)9798350329223

Balancing robustness and computational efficiency in machine learning models is challenging, especially in settings with limited resources like mobile and IoT devices. This study introduces Adaptive and Localized Adversarial Training (ALAT), an optimization approach that balances these competing needs. ALAT combines generalized models with localized adversarial perturbations and adaptive data augmentation. As a result, the model strengthens its weak points without needing to explore all possible adversarial threats, saving computational effort Our data shows that ALAT-trained models perform robustly with less computational cost compared to traditional adversarial training methods. This adaptability makes ALAT suitable for various machine learning architectures and particularly valuable in resource-constrained settings requiring resilience to adversarial threats. © 2023 IEEE.

关键词： Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Trust and Risk Based on Access Control Model in Social Internet of Things

引用

International Journal of Network Security 2023年第5期25卷 736-744页

作者： Zhang, Hongbin Liu, Jian Zhao, Dongmei Liu, Bin Wang, Yanmei Fan, Fan School of Information Science and Engineering Hebei University of Science and Technology Shijiazhuang050000 China Hebei Key Laboratory of Network and Information Security Hebei Normal University Shijiazhuang050024 China School of Economics and Management Hebei University of Science and Technology Shijiazhuang050000 China Research Center of Big Data and Social Computing Hebei University of Scienc and Technology Shijiazhuang050000 China Hebei Geological Worker’s University Shijiazhuang050000 China

Integrating social networks and the Internet of Things has formed an emerging field of Social Internet of Things (IoT). Aiming to address the shortcomings of traditional TBAC (Trust based on Access Control) and ABAC (Attribute based on Access Control) in dynamic authorization and security, this paper proposes a hybrid access control model based on trust and risk for SIoT. The model supports social relationship-based trust assessment and risk prediction between SIoT nodes and integrates trust and risk values into SIoT security values through reasonable weight allocation. Furthermore, by simulating the safety requirements of different scenarios, it was verified that the model was efficient and more secure than the current models. © (2023). All Rights Reserved.

关键词： Authorization

来源：评论

学校读者我要写书评

暂无评论

Learning in a digital fabrication course on building tangible artefacts

Learning in a digital fabrication course on building tangibl...

引用

2024 International Design Society Conference, Design 2024

作者： Nanjappan, Vijayakumar Georgiev, Georgi V. Casakin, Hernan Soomro, Sohail Ahmed Center for Ubiquitous Computing University of Oulu Finland School of Computer Science and Information Technology University College Cork Ireland Ariel University Israel Sukkur IBA University Pakistan

This paper examines how students' ideas evolve into physical prototypes within a digital fabrication design course. Examining the materials used, customization approaches, iterations, and team dynamics of 26 student projects reveals interplays between ideas, available tools, materials and constraints. Findings show the predominance of techniques, design preferences, concept refinement, and teamwork challenges. The implications highlight the value of hands-on iteration for alignment with reality and the need to support collaboration skills alongside technical prototype development. © 2024 Proceedings of the Design Society. All rights reserved.

关键词： 3D printing

来源：评论

学校读者我要写书评

暂无评论

UMixer: A Novel U-shaped Convolutional Mixer for Multi-scale Feature Fusion in Medical Image Segmentation 16th

UMixer: A Novel U-shaped Convolutional Mixer for Multi-scal...

引用

16th Chinese Conference on Biometric Recognition, CCBR 2022

作者： Su, Yongxin Huang, Hongbo Song, Zun Lin, Lei Liu, Jinhan Mechanical Electrical Engineering School Beijing Information Science and Technology University Beijing China Computer School Beijing Information Science and Technology University Beijing China School of Economics and Management Beijing Information Science and Technology University Beijing China Institute of Computing Intelligence Beijing Information Science and Technology University Beijing China

ISBN: (纸本)9783031202322

Medical image segmentation plays a critical role in assisting diagnosis and prognosis. Since the transformer was first introduced to the field, the neural network structure has experienced a transition from ConvNet into Transformer. However, some redesigned ConvNets in recent works show astonishing effects, which outperforms classic elaborate ConvNets, and even complicated transformers. Inspired by these works, we introduced large-kernel convolutions to improve the ConvNets in capturing the long-range dependency. Cooperated with a novel multi-scale feature fusion method, we proposed a U-shaped convolutional structure, dubbed UMixer, which effectively integrates shallow spatial information with deep semantic information and high-resolution detailed information with low-resolution global information. Without any attention mechanism and pre-training on large datasets, UMixer achieves more accurate segmentation results than traditional ConvNets and Transformers on the Synapse dataset. Experiments demonstrate the effectiveness of this multi-scale feature fusion structure and its capability in modeling long-range dependency. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Large dataset

来源：评论

学校读者我要写书评

暂无评论

Sequential Topic Selection Model with Latent Variable for Topic-Grounded Dialogue

Sequential Topic Selection Model with Latent Variable for To...

引用

2022 Findings of the Association for Computational Linguistics: EMNLP 2022

作者： Wen, Xiaofei Wei, Wei Mao, Xian-Ling Cognitive Computing and Intelligent Information Processing Laboratory School of Computer Science and Technology Huazhong University of Science and Technology China Department of Computer Science and Technology Beijing Institute of Technology China

Recently, topic-grounded dialogue system has attracted significant attention due to its effectiveness in predicting the next topic to yield better responses via the historical context and given topic sequence. However, almost all existing topic prediction solutions focus on only the current conversation and corresponding topic sequence to predict the next conversation topic, without exploiting other topic-guided conversations which may contain relevant topic-transitions to current conversation. To address the problem, in this paper we propose a novel approach, named Sequential Global Topic Attention (SGTA) to exploit topic transition over all conversations in a subtle way for better modeling post-to-response topic-transition and guiding the response generation to the current conversation. Specifically, we introduce a latent space modeled as a Multivariate Skew-Normal distribution with hybrid kernel functions to flexibly integrate the global-level information with sequence-level information, and predict the topic based on the distribution sampling results. We also leverage a topic-aware prior-posterior approach for secondary selection of predicted topics, which is utilized to optimize the response generation task. Extensive experiments demonstrate that our model outperforms competitive baselines on prediction and generation tasks. © 2022 Association for Computational Linguistics.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

An identity-based message authentication scheme for Internet of vehicles 8

An identity-based message authentication scheme for Internet...

引用

8th IEEE International Conference on Big Data Security on Cloud, 8th IEEE International Conference on High Performance and Smart computing, and 8th IEEE International Conference on Intelligent Data and Security, BigDataSecurity/HPSC/IDS 2022

作者： Yan, Chenyang Wang, Hongshuo Yan, Songsong QILU Institute of Technology School of Computing Science and Information Engineering Jinan China The University of Nottingham Nottingham United Kingdom School of Computing Science and Information Engineering Jinan China

ISBN: (纸本)9781665480697

The Internet of Vehicles realizes information sharing through the interconnection between entities such as vehicles and vehicles, vehicles and roads. As the computing tasks of vehicles become more and more complex, the data load of traditional cloud computing centers also increases. If a large amount of data generated by the vehicle is all uploaded to the cloud server, and the data is processed by the cloud computing center, it will increase the burden on the cloud server. In order to reduce the request time in the vehicle and reduce the data transmission delay, edge computing is introduced, and an identity-based vehicle networking message authentication scheme in the edge computing scenario is designed to realize the rational use of computing and storage resources. The scheme satisfies unlinkability, unforgeability, identity privacy protection, and anti-replay attack. © 2022 IEEE.

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

Dense Visible Watermark Removal with Progressive Feature Propagation 12th

Dense Visible Watermark Removal with Progressive Feature Pr...

引用

12th International Conference on Image and Graphics, ICIG 2023

作者： Ren, Chunchi Zhao, Jiangfeng He, Hongjie Chen, Fan School of Information Science and Technology Southwest Jiaotong University Chengdu611756 China School of Computing and Artificial Intelligence Southwest Jiaotong University Chengdu611756 China

ISBN: (纸本)9783031463167

Visible watermark removal is an important task to ensure the robustness of visible watermarks. However, existing techniques are unable to efficiently remove dense visible watermarks in stock content market images. To address this issue, we propose a novel solution called the Progressive Visible Watermark Removal Network (PVWRNet) that achieves efficient and effective dense visible watermark removal. Our approach utilizes a cross-combination structure in the backbone module to balance dense visible watermark spatial and contextual features. Furthermore, progressive modules are incorporated to multi-fuse shallow features of the backbone module, leading to significant improvement in image restoration metric scores and visual quality. To enhance local features of the backbone module, Reinforced Dual Attention Modules (RDAM) are introduced, which reweight the local features. Experimental results demonstrate that our proposed PVWRNet outperforms state-of-the-art approaches in the dense visible watermark dataset, especially in high definition images. In conclusion, the proposed method provides a potent and efficient solution for removing dense visible watermarks in stock content market images, and the use of cross-combination structure, progressive modules, and reinforced dual attention modules enables PVWRNet to outperform existing techniques in solving the challenging dense visible watermark removal problem. The code is available at: https://***/HyuuHyuuNeko/PVWRNet. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Diabetes Prediction using Enhanced and Optimized DRL-CNN Approaches 21

Diabetes Prediction using Enhanced and Optimized DRL-CNN App...

引用

21st International Learning and technology Conference, L and T 2024

作者： Usha Nandhini, A. Dharmarajan, K. Sharma, Bhisham Chowdhury, Subrata Ben Dhaou, Imed Vistas School of Computing Chennai India Vistas Department of Information Technology Chennai India Chitkara University Institute of Engineering and Technology Chitkara University Department of Computer Science and Engineering Punjab India Sreenivasa Institute of Technology and Management Studies Department of Computer Science and Engineering Andra Pradesh Chittoor India Dar Al-Hekma University Hekma School of Engineering Computing and Design Department of Computer Science Jeddah Saudi Arabia

ISBN: (纸本)9798350393569

This study develops an advanced automated prediction system using Machine Learning (ML) techniques to identify diabetes early. The research employs the WBSMOTE method for data preprocessing, addresses class imbalances with the WB-SMOTE approach, and introduces two novel classification methods, DRL-CNN and DRL-OCNN. Reliable datasets from sources like The Data World repository, PIMA Indian Diabetes (PIDD) dataset, and Real Time (RT) Diabetes Dataset are used, with an emphasis on data consistency and feature relevance. If we examine these datasets, it contains general attributes and no internal specific attributes related to diabetes is used. So, prediction results will be what type of diabetes we won't get any specific characteristics and risk factors. Evaluation is based on Accuracy, Precision, and Recall criteria. The DRL-OCNN model, integrating SGD optimization and Delta Rule Learning, outperforms other ML models with a accuracy of 96%, precision of 0.93 and recall of 0.91, demonstrating its potential for accurate early diabetes identification and classification. © 2024 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Flexible Global Aggregation and Dynamic Client Selection for Federated Learning in Internet of Vehicles

引用

Computers, Materials & Continua 2023年第11期77卷 1739-1757页

作者： Tariq Qayyum Zouheir Trabelsi Asadullah Tariq Muhammad Ali Kadhim Hayawi Irfan Ud Din College of Information Technology United Arab Emirates UniversityAl Ain17551UAE Department of Computing School of EECSNational University of Sciences and TechnologyIslamabad44000Pakistan College of Technological Innovation Zayed UniversityAbu Dhabi144534UAE Department of Computer Science The Superior UniversityLahore54000Pakistan

Federated Learning(FL)enables collaborative and privacy-preserving training of machine learning models within the Internet of Vehicles(IoV)*** FL effectively tackles privacy concerns,it also imposes significant resource *** traditional FL,trained models are transmitted to a central server for global aggregation,typically in the *** approach often leads to network congestion and bandwidth limitations when numerous devices communicate with the same *** need for Flexible Global Aggregation and Dynamic Client Selection in FL for the IoV arises from the inherent characteristics of IoV *** include diverse and distributed data sources,varying data quality,and limited communication *** employing dynamic client selection,we can prioritize relevant and high-quality data sources,enhancing model *** address this issue,we propose an FL framework that selects global aggregation nodes dynamically rather than a single fixed *** global aggregation ensures efficient utilization of limited network resources while accommodating the dynamic nature of IoV data *** approach optimizes both model performance and resource allocation,making FL in IoV more effective and *** selection of the global aggregation node is based on workload and communication speed ***,our framework overcomes the constraints associated with network,computational,and energy resources in the IoV environment by implementing a client selection algorithm that dynamically adjusts participants according to predefined *** approach surpasses Federated Averaging(FedAvg)and Hierarchical FL(HFL)regarding energy consumption,delay,and accuracy,yielding superior results.

关键词： IoT Federated Learning sensors IoV OMNeT++ edge computing

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：