检索结果-内蒙古大学图书馆

Adaptive Central Acceleration With Variance Control for Robust Federated Optimization in Ubiquitous Intelligence

IEEE Internet of Things Journal 2025年第12期12卷 21379-21393页

作者： Zhao, Lei Lu, Wu-Sheng Cai, Lin University of Victoria Dept. of Electrical and Computer Engineering VictoriaBCV8P 5C2 Canada

Federated Learning (FL) in Intelligent Internet of Things (IIoT) environments faces critical challenges, including sparse client participation, non-IID local data distributions, and unreliable communication, which lead to slow convergence and high variance in global updates. To address these issues, we propose Adaptive Central Federated Momentum Optimization (ACFMO), an optimization framework that enhances FL efficiency and stability under constrained participation. ACFMO integrates an adaptive central acceleration mechanism that dynamically adjusts momentum updates based on real-time client availability, preventing instability and ensuring smoother global model updates. Additionally, a variance-controlled local updating strategy refines client contributions, mitigating high variance caused by infrequent and heterogeneous updates. Extensive experiments across diverse FL scenarios demonstrate that ACFMO significantly accelerates convergence, reduces communication overhead, and improves model stability compared to state-of-the-art FL methods, making it particularly well-suited for real-world IIoT deployments where network and computational resources are constrained. © 2025 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Smart Gadgets Cockpit Using IoT 1

Smart Gadgets Cockpit Using IoT

引用

1st IEEE International Conference on Computing and Data Science, ICCDS 2024

作者： Eapen, Maya Mercy, H. Beulah Jayakumari, R. Lavanya Mala, A. Indumathi, S. Pooja, S. Jerusalem College oif Engineering Dept. of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350365337

The Smart Device Cockpit collects data from IoT enabled household appliances like Air conditioner, refrigerators, microwave ovens, and washing machines. Once transmitted to the Smart Device Cockpit portal, the data undergoes analysis and is then presented in the form of both usage data and error data. This platform facilitates a thorough comprehension of the utilization patterns of these household devices and offers insights into any possible errors or malfunctions that may occur. In essence, the Smart Device Cockpit serves as a powerful tool for homeowners to not only monitor and manage their smart appliances but also to make informed decisions about usage optimization and timely maintenance. © 2024 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Hybrid Residual Blocks - Transformers Improving Cardiac MRI Segmentation in Deep Learning

Hybrid Residual Blocks - Transformers Improving Cardiac MRI ...

引用

2023 IEEE International Conference on Communications, Information, Electronic and Energy systems, CIEES 2023

作者： Naser, Maysaa Abd Ulkareem Al-Asadi, Abbas H. Hassin University of Basrah Dept. of Computer Information Systems Basrah Iraq Kufa University Dept. of Computer Science Kufa Iraq

ISBN: (纸本)9798350336917

Cardiovascular disease is one of the primary causes of mortality and disability. By assisting in early diagnosis, segmentation techniques based on deep learning might help to lessen their severity, although extremely high levels of accuracy are required. Introduce a novel and flexible architecture that raises the bar for semantic segmentation accuracy in this study. Our contributions focus on the merging of strong self-attention mechanisms obtained from Transformer with traditional convolutional neural network (CNN) architecture. This hybrid model demonstrates improved feature representation, contextual understanding, and pixel-wise predictions, setting new standards in semantic segmentation tasks. the average dice coefficients for ACDC dataset by our model, AVG DICE 92.52%, RV 90.77%, MYO 90.00%, LV 96.2S% © 2023 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Approach for Multi-Class Identification from Domestic Violence Online Posts 15

Deep Learning Approach for Multi-Class Identification from D...

引用

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Yadav, Abhay Pravin Saraiya, Jigar Vishanji Garud, Chaitanya Vasant Pate, Shivam Rajendra Thakur, Payel Dept. of Computer Engineering Pillai College of Engineering Maharashtra Panvel410206 India

ISBN: (纸本)9798331300579

Deep Learning for Multi-Class Identification From Domestic Violence Online Posts involves training a neural network to recognize patterns and features in text that are indicative of domestic violence. The aim is to develop, train and evaluate a multiclass classification model on a dataset of online posts so that model can identify and classify domestic violence-related content from online posts. By conducting various analysis tests, models will predict accurately and distinguish between different types of domestic violence, in order to better understand the prevalence and specific forms of abuse occurring within online communities. The model will be trained on a dataset of online posts and will be evaluated on its ability to accurately classify new, unseen data. The audio, selected text, image and will be classified to display the results. The outcome of this project will be a deep learning model that can effectively identify and classify different forms of domestic violence in online posts. © Grenze Scientific Society, 2024.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

SFNet: Synergistic Fusion Network for Text Localization in Natural Scene Images 3

SFNet: Synergistic Fusion Network for Text Localization in N...

引用

3rd International Conference for Innovation in Technology, INOCON 2024

作者： Suman, Suman Champa, H.N. University of Visvesvaraya College of Engineering Dept. of Computer Science and Engineering Bangalore India

ISBN: (纸本)9798350381931

Prior methodologies for the detection of scene text have demonstrated promising outcomes across many evaluation metrics. Nevertheless, despite the utilization of deep neural network models, their effectiveness is sometimes limited in complicated situations due to the intricate interplay of several stages and components within the pipelines, which ultimately governs overall performance. This study presents a straightforward yet robust methodology for identifying textual content in practical environments. Phrases or text lines may be predicted in images of varying orientations and quadrilateral forms using the suggested pipeline's single neural network. This eliminates the need for superfluous intermediary steps such as candidate aggregation and word division. Due to the streamlined nature of our SFNet(Synergistic Fusion Network), our primary emphasis is directed at the development of loss functions and neural network design. The suggested approach has been evaluated using a well-established dataset including ICDAR 2015. The results of these experiments demonstrate that the new technique outperforms existing techniques in terms of both accuracy and efficiency. The proposed methodology demonstrates a performance metric, namely an F-score of 0.89 while operating at a frame rate of 13.2 frames per second and a resolution of 720p using the ICDAR 2015 dataset. © 2024 IEEE.

关键词： Pipelines

来源：评论

学校读者我要写书评

暂无评论

Unveiling Insights: Vision-Based Data Mining Analysis of Webpages in Transition from HTML 3 to HTML 5 5

Unveiling Insights: Vision-Based Data Mining Analysis of Web...

引用

5th IEEE International Conference on Computing, Power, and Communication Technologies, IC2PCT 2024

作者： Chaudhary, Meena Pooja Chandwani, Gunjan Manav Rachna University Dept. of Computer Science and Technology Faridabad India Galgotias University Dept. of Computer Science and Engineering Greater Noida India

ISBN: (纸本)9798350383522

Deep web material is accessed with web database requests, and returned information is packaged into dynamically generated web pages known as deep Web pages. Due to their intricate structure, it is tedious task to retrieve organized data from deep web pages. A great many approaches to address this problem are yet to be proposed, but all of them have inherent drawbacks because they are based on the website and the programming language. The contents of Sites are always shown for users to search daily, like the common two-dimensional media. This motivates us to search for an alternative way to extract deep Web information, by using some interesting popular visual features from deep Webpages, to address the limitations of past works. A novel blog vision approach is proposed in this paper to overcome these issues. This technique uses visual features in deep web pages specifically to conduct deep web data extraction. This paper also recommends a new appraisal measure to capture the required human activity to make maximum mining. The proposed methodology is highly efficient in the deep network data extraction, according to our studies with a large set of web-based data bases. © 2024 IEEE.

关键词： Data mining HTML 3 HTML5 Vision based technique Web Content Mining

来源：评论

学校读者我要写书评

暂无评论

Image Segmentation based Approach for Skin Disease Detection and Classification using Machine Learning Algorithms 2

Image Segmentation based Approach for Skin Disease Detection...

引用

2nd IEEE International Conference on Integrated Circuits and Communication systems, ICICACS 2024

作者： Sany, Nazmul Hasan Chandra Shill, Pintu Khulna University of Engineering & Technology Dept. of Computer Science and Engineering Khulna Bangladesh

ISBN: (纸本)9798350317558

Skin disorders are ubiquitous, stemming from causes including fungus, germs, allergies, and viruses. Although laser and photonics medical technologies have revolutionised rapid and precise diagnosis, their exorbitant prices restrict accessibility. To address this, this paper describes an image processing strategy for dermatological screening and procedure that comprises obtaining digital photographs of damaged skin patches and applying image processing to detect the disease type. Moreover, when conducted manually by medical specialists, identifying skin illnesses is time-intensive and subjective. As a consequence, both patients and physicians want automated skin disease prediction, which makes the treatment plan faster. This cost-effective approach depends on a basic setup: a camera and a computer. Operating on colour photos, a solution combines resizing for feature extraction using a pertained convolutional neural network, followed by classification using a multiclass SVM. The technology efficiently recognises four separate skin illnesses with an astonishing 90.8% accuracy, giving users complete findings spanning disease kind, extent, and severity. © 2024 IEEE.

关键词： Decision Tree Image Processing K-NN Machine Learning Skin Disorder SVM

来源：评论

学校读者我要写书评

暂无评论

Unravel News: An Efficient Summarization Approach

Unravel News: An Efficient Summarization Approach

引用

2024 IEEE International Conference on Computing, Applications and systems, COMPAS 2024

作者： Saha, Ankan Al Shafi, Abdullah Khulna University of Engineering and Technology Dept. Computer Science and Engineering Khulna Bangladesh

ISBN: (纸本)9798331529765

The process of automatically creating a concise and informative summary of a news piece is known as news summarizing. This can be a useful tool for people who want to quickly get the gist of a news story without having to read the entire article. Extractive and abstractive summarization are the two basic methods used to summarize news. The process of extractive summarization entails locating and separating the key phrases from the source text. In this paper, sentence weighting is used for extractive summarization. The words in the headlines are given ten times more emphasis than the other words in the news article. The final summarization is done by rearranging the sentences as per the sequence in the main news article. The paper concludes by discussing the future of news summarization, which is an active area of research. © 2024 IEEE.

关键词： Abstractive Summary Extractive Summary ROUGE Sentence Rank Algorithm Stemming Text Rank Algorithm

来源：评论

学校读者我要写书评

暂无评论

A Fusion-Based Approach to Multiclass Retinal Disease Classification: Transfer Learning Strategy with CBAM Integration 2

A Fusion-Based Approach to Multiclass Retinal Disease Classi...

引用

2nd International Conference on Power, Electrical, Electronics and Industrial Applications, PEEIACON 2024

作者： Nath, Bijoy Deb Afroge, Shyla Rajshahi University of Engineering & Technology Dept. of Computer Science & Engineering Rajshahi Bangladesh

ISBN: (纸本)9798331517984

Eye diseases pose significant global public health challenges, often resulting to irreversible blindness if untreated. Early, precise diagnosis of these disorders is critical. In this paper, we address this crucial issue using transfer learning methods for the categorization of multiple eye disorders. The proposed architecture features a fusion-based technique that combines EfficientNetB0 and MobileNetV3Large convolutional neural networks (CNNs) for feature extraction and incorporates Convolutional Block Attention Module (CBAM), which dynamically focuses on significant regions of the images, enhancing the model's discriminative capability. Our proposed model achieves outstanding performance on two separate datasets: 97.85% accuracy on the Eye Disease Classification (EDC) dataset and 91.96% on the Ocular Disease Intelligent Recognition (ODIR) dataset, surpassing existing state-of-the-art approaches in ocular disease classification. By developing the proposed model, we intend to contribute to the creation of more effective diagnostic tools and eventually enhance patient outcomes in ophthalmology. © 2024 IEEE.

关键词： Ophthalmology

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Model for Weather Prediction 1

Deep Learning Model for Weather Prediction

引用

1st International Conference on Computing, Sciences and Communications, ICCSC 2024

作者： Singh, Shubhanshi Tyagi, Neha Amity School of Engineering and Technology Dept. Computer Science and Engineering Noida India

ISBN: (纸本)9798350353648

Weather forecasting predicts atmospheric conditions at a particular time and place. It is an important task with a wide range of applications, including agriculture, transportation, and disaster prevention. Traditional weather forecasting methods rely on physical models of the atmosphere, which can be complex and computationally expensive. In recent years, deep learning techniques are effective for weather forecasting. Deep learning models can learn complex patterns from data, which can be used to predict future weather conditions accurately. This project aims to create a deep-learning model for forecasting weather. A huge dataset of historical meteorological data, including radar, satellite, and surface measurements, will be used to train the model. Numerous meteorological factors, such as temperature, precipitation, and wind speed, will be predicted by the model. We will compare the deep learning model's performance with conventional weather forecasting techniques. Creating a deep learning model that is more accurate and effective than current approaches is the project's main objective. The initiative may result in improved weather forecasts that are prompt and accurate, better agricultural, transportation, and disaster prevention decision-making, and less money lost because of weather-related incidents. Possible obstacles for the project are gathering and purifying a sizable dataset of meteorological data, using this data to train a deep learning model, and then comparing the model's output with more conventional techniques. © 2024 IEEE.

关键词： Wind forecasting

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：