检索结果-内蒙古大学图书馆

Predictive analysis-based load balancing and fault tolerance in fog computing environment

Cluster Computing 2025年第5期28卷 1-16页

作者： Kashyap, Vijaita Ahuja, Rakesh Kumar, Ashok Chitkara University Institute of Engineering & Technology Chitkara University Punjab Rajpura India Department of Computer Science and Engineering Chandigarh Engineering College Chandigarh Group of Colleges Punjab Jhanjeri India

Fog computing can process data in real-time, which may minimize network congestion and delays. Load balancing and fault tolerance are critical matrices for allocating resources in a dynamic fog computing environment. They ensure that processes can continue to function even when something goes wrong, and that work is distributed properly. To optimize resource allocation in fog computing settings, this study presents a novel technique, Predictive analysis of Load balancing and Fault Tolerance (PreLoFT). The technique dynamically distributes workloads among fog nodes and forecasts resource requirements using predictive analysis. The method incorporates improved task-splitting to handle resource-intensive tasks. Additionally, it enhances fault tolerance by reallocating workloads when a node fails, enhancing the overall reliability and robustness of the system. PreLoFT’s effectiveness is evaluated and compared to more advanced algorithms like the Honeybee Foraging Algorithm (HFA), Genetic Algorithm (GA), and Load-aware Energy-efficient Scheduling (LECS), as well as traditional methods like Round-Robin (RR) and Least Loaded (LL). PreLoFT considerably enhances several metrics, such as execution time, resource use, load balancing efficiency, fault tolerance, latency, and energy consumption. It is notable for its ability to provide greater resource utilization (90%) by proactive workload management, which guarantees that nodes are not over-utilized and are functioning near their capacity. While high resource utilization maximizes performance and reduces the need to offload tasks to higher-latency cloud resources. The method also maintains energy efficiency by preventing idle resources and minimizing task migrations. The findings demonstrate that PreLoFT outperforms traditional load balancing techniques by providing faster execution times, lower latency, and better energy management, making it a promising approach for real-time fog computing environments. © The Author(s), und

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Sample-based relationship for assisting diagnosis of pneumonia in medical care

引用

Multimedia Tools and Applications 2025年第1期84卷 267-288页

作者： Chen, Hongkang Lu, Huijuan Zhu, Wenjie Zhou, Ye Yao, Yudong Wang, Renfeng College of Information Engineering China Jiliang University Zhejiang Hangzhou China Drore Information and Technology Co. LTD Zhejiang Hangzhou China Department of Electrical and Computer Engineering Stevens Institute of Technology HobokenNJ United States College of Computer Science and Technology Zhejiang University Zhejiang Hangzhou China

Deep neural networks have succeeded in learning balanced and imbalanced data in the field of pneumonia diagnosis. However, both require separate model designs in their respective domains. The pneumonia recognition methods have poor transfer performance between the two types of image distribution data. This paper proposes BeFormer, which focuses on sample relationships. BeFormer introduces two stages, feature extraction and establishing sample relationships, to solve the problem. Specifically, in the feature extraction stage, a new backbone network called PANet based on Multi-Scale Frequency Domain Attention (FPSA) is proposed. In establishing sample relationships, Deformable Transformer is used in parallel to model the relationships between partial regions of interest among the same and different samples in the pneumonia dataset. These two modules are connected in series to form BeFormer. Without making any changes to the model, this paper conducted experiments using BeFormer on both balanced and imbalanced pneumonia datasets. As compared to traditional pneumonia recognition models, the proposed method achieved significant improvements in the two different data distributions, demonstrating outstanding transfer performance. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Cyber-Physical Systems and IoT Intrusion Detection: A Hybrid Deep Learning Perspective 3

Cyber-Physical Systems and IoT Intrusion Detection: A Hybrid...

引用

3rd International Conference on Electronics and Renewable Systems, ICEARS 2025

作者： Prakash, S. Jaya Devi, D. Muthukumaran, N. Romila, R. Idhaya Engineering College for Women Department of Computer Science and Engineering Tamilnadu Chinnasalem India Maha Barathi Engineering College Department of Computer Science and Engineering Tamilnadu Chinnasalem India A.R. Engineering College Department of Computer Science and Engineering Tamilnadu Villupuram India Sri Muthukumaran Institute of Technology Department of Information Technology Tamilnadu Chennai India

ISBN: (纸本)9798331509675

Several vital resources are increasingly being protected by cyber-physical systems (CPSs), makes the detection of incidents on these systems critical. CPSs along with other domains, such as the Internet of Things (IoT), frequently use machine learning (ML) and deep learning (DL) approaches to combat intrusion detection. Existing approaches, however, frequently favor the deployment of advanced detection models over their usefulness in actual operations. This research work develops a hybrid intrusion detection technique by combining Convolutional Neural Networks (CNN) with Long Short-Term Memory (LSTM). This work analyzes the experimental results to LSTM and CNN to develop an effective comparative approach. The experiments are conducted on the two datasets. The efficiency indicators examined are intrusion detection results such as accuracy, precision, recall, and f1score. This comparison indicates that the proposed hybrid algorithm is the preferable approach. © 2025 IEEE.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

A comprehensive review on detection of depression through social media analysis 4th

A comprehensive review on detection of depression through so...

引用

4th International Conference on Computational Methods in science and technology, ICCMST 2024

作者： Dogra, Shriya Kaur, Parneet Sharma, Tejpal Kumari, Nisha Department of Computer Science and Engineering Chandigarh Engineering College-CGC Landran India Department of Computer Science and Engineering Chitkara University Institute of Engineering and Technology Punjab India

ISBN: (纸本)9781032911571

Social media platforms like Instagram, Twitter, and Facebook have completely changed our world. People today exhibit a kind of digital character and are more linked than ever. While social media undoubtedly offers many noteworthy advantages, there are also unavoidable drawbacks. Later Research has demonstrated a link between elevated use of social media platforms and heightened levels of depression. The goal of the current study is to study the existing machine learning approaches to identify potentially sorrowful Twitter users based on their tweets and network behaviour. To accomplish this, we assessed classifiers that identify whether or not a person is depressed based on information extracted from their tweets and network behaviour. It is observed that as a greater number of features are incorporated, the F-score and accuracy in classifying depressed persons also improve. This technique is a predictive, data-driven tool for early mental health diagnosis of depression or any type of psychiatric ailments. The main contribution of the article is the examination of the traits and how they affect the ability to gauge depression severity. © 2025 the Author(s).

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

Deep Learning for Cross-Domain Few-Shot Visual Recognition: A Survey

引用

ACM Computing Surveys 2025年第8期57卷 1-37页

作者： Xu, Huali Zhi, Shuaifeng Sun, Shuzhou Patel, Vishal Liu, Li College of Computer Science Nankai University Tianjin China College of Electronic Science and Technology National University of Defense Technology Changsha China Faculty of Information Technology and Electrical Engineering University of Oulu Oulu Finland Johns Hopkins University Baltimore United States

While deep learning excels in computer vision tasks with abundant labeled data, its performance diminishes significantly in scenarios with limited labeled samples. To address this, Few-shot learning (FSL) enables models to perform the target tasks with very few labeled examples by leveraging prior knowledge from related tasks. However, traditional FSL assumes that both the related and target tasks come from the same domain, which is a restrictive assumption in many real-world scenarios where domain differences are common. To overcome this limitation, Cross-domain few-shot learning (CDFSL) has gained attention, as it allows source and target data to come from different domains and label spaces. This article presents the first comprehensive review of Cross-domain Few-shot Learning (CDFSL), a field that has received less attention compared to traditional FSL due to its unique challenges. We aim at providing both a position article and a tutorial for researchers, covering key problems, existing methods, and future research directions. The review begins with a formal definition of CDFSL, outlining its core challenges, followed by a systematic analysis of current approaches, organized under a clear taxonomy. Finally, we discuss promising future directions in terms of problem setups, applications, and theoretical advancements. © 2025 Copyright held by the owner/author(s).

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Neuroadaptive Containment Control for Nonlinear Multiagent Systems With Input Saturation: An Event-Triggered Communication Approach

引用

IEEE Transactions on Systems, Man, and Cybernetics: Systems 2025年第5期55卷 3163-3173页

作者： Wang, Xin Zhang, Sen Li, Huaqing Zhang, Wei Li, Hongyi Huang, Tingwen Southwest University College of Electronic and Information Engineering Chongqing400715 China Shenzhen University of Advanced Technology Faculty of Computer Science and Control Engineering Shenzhen518055 China

In this article, a neuroadaptive event-triggered containment control strategy combined with the dynamic surface control (DSC) approach is proposed for nonlinear multiagent systems (MASs) with input saturation. Based on the event-triggered communication mechanisms, the updates of neural network weight and controllers are implemented solely under triggering conditions of violation, which markedly reduces unnecessary communication resources and minimizes inefficient control costs compared with the traditional control method. Radial basis function neural networks (RBF NNs) are employed to handle the nonlinear uncertainties of MASs. Simultaneously, an adaptive compensatory mechanism is incorporated within the backstepping process to address the nonlinear effect posed by input saturation. Additionally, we demonstrate the system stability by extending the Lyapunov theorem to jump and continuous scenarios while excluding Zeno behavior, realizing that all followers can enter the convex hull constructed by leaders. Finally, the effectiveness of the proposed methodology is verified through application simulations. © 2013 IEEE.

关键词： Radial basis function networks

来源：评论

学校读者我要写书评

暂无评论

Climate Geoscience–Based Disaster Management in Healthcare Analysis by Markov Q-Transfer Adversarial with Cloud Computing

引用

Remote Sensing in Earth Systems sciences 2025年第2期8卷 474-483页

作者： Srinivas, Jhade Rebecca, B. Bhargavi, Kovvuri N Swetha, Velma Reddy Spandana, S. Degala, Divya Priya Department of Computer Science and Engineering-Data Science KG Reddy College of Engineering and Technology Moinabad Telangana Hyderabad India Computer Science and Engineering Marri Laxman Reddy Institute of Technology and Management Telangana Dundigal 500043 India Department of CSE Aditya University AP Surampalem India Department of Computer Science and Engineering (AI-ML) Malla Reddy Engineering College Hyderabad India Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Bowrampet Telangana Hyderabad India Department of Computer Science and Engineering MLR Institute of Technology Hyderabad India

Analysis and reaction to natural disasters have made extensive use of deep learning methods using semantic segmentation networks. These implementations’ foundation is based on convolutional neural networks (CNNs), which are capable of precisely identifying and locating the relevant areas of interest within satellite imagery or other types of remote sensing data. This helps with rescue planning, restoration efforts, and disaster evaluation. Difficulties in the provision of healthcare services, due to natural disasters, are made worse by the existing problems prevalent in most healthcare systems. The aim of this research is to propose a novel technique in climate change–based disaster management with their public healthcare analysis using cloud computing with deep learning techniques. In this research, the cloud network has been used in LiDAR image collection for climate change–based disaster-struck regions and their public healthcare analysis. This collected image has been classified and optimized using a deep Markov Q-transfer adversarial neural network with binary grey whale swarm optimization. Experimental analysis has been carried out for various disaster management–based public healthcare datasets in terms of classification accuracy, specificity, NSE, sensitivity, and F-1 score. The proposed technique attained a classification accuracy of 97%, F-1 score of 94%, specificity of 96%, sensitivity of 95%, and NSE of 57%. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025.

关键词： Adversarial neural network Cloud computing Disaster management LiDAR image Whale swarm optimization

来源：评论

学校读者我要写书评

暂无评论

Breast cancer diagnosis using machine learning model 4th

Breast cancer diagnosis using machine learning model

引用

4th International Conference on Computational Methods in science and technology, ICCMST 2024

作者： Arthi, T.S. Partheeban, N. Singh, Bhoomika Singh, Aparajitha Tripathi, Nikhil Department of Computer Science and Engineering Galgotias College of Engineering & Technology Greater Noida India Department of Computer Science and Engineering Galgotias University Greater Noida India

ISBN: (纸本)9781032911571

Early and accurate detection of breast cancer, particularly Invasive Ductal Carcinoma (IDC), is critical for improving patient outcomes. Traditional diagnostic methods like histopathology and mammography have limitations in terms of time, cost, and potential subjectivity. This study proposes a lightweight Machine Learning (ML) model for IDC classification using digital whole-mount breast cancer slide images. The model aims to achieve high accuracy while being suitable for deployment in resource-constrained environments. We evaluate our model on a publicly available breast histopathology dataset and compare its performance with other Deep Learning (DL) models, focusing on both accuracy and computational efficiency. The results demonstrate that our model achieves a high classification accuracy (around 89.3 percent) with minimal processing power and memory requirements. Furthermore, successful deployment on a Raspberry Pi embedded system validates its potential for real-world application in resource-limited settings, potentially leading to improved breast cancer diagnosis accessibility. Index Terms—Bi-directional Long Short-Term Memory (BiLSTM), Convolutional Neural Network (CNN), Deep Learning (DL), Invasive Ductal Carcinoma (IDC), Multilayer Perceptron (MLP), Raspberry Pi, Tiny Machine Learning (TinyML). © 2025 the Author(s).

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A combinatorial algorithm for finding straight skeleton of a digital object

引用

Multimedia Tools and Applications 2025年 1-21页

作者： Maity, Anukul Dutt, Mousumi Biswas, Arindam Department of Computer Science and Engineering Narula Institute of Technology 81 Nilgunj Road Agarpara West Bengal Kolkata700109 India Department of Computer Science and Engineering St. Thomas’ College of Engineering and Technology 4 Diamond Harbour Road Kidderpore West Bengal Kolkata700023 India Department of Information Technology Indian Institute of Engineering Science and Technology Botanical Garden Road Shibpur West Bengal Howrah711103 India

The paper presents a combinatorial algorithm to find the straight skeleton of the inner isothetic cover of a digital object imposed on a uniform background grid. The isothetic polygon (orthogonal polygon) tightly inscribes the given digital object. The algorithm stated here finds the straight skeleton of any isothetic polygon (simple or non-simple) in O(nglogng) time in a single traversal by applying combinatorial rules, where n is the number of pixels on the boundary of the digital object and g is the grid size on which the digital object is imposed. To find the straight skeleton of non-simple orthogonal polygon, the polygon is divided into sub-parts which are a set of simple orthogonal polygons. The straight skeleton is generated for each of the sub-parts and the corresponding results are merged. The orthogonal polygon containing holes is cut at each holes and the polygon becomes hole-free. The algorithm to obtain the straight skeleton is applied on it. The disconnected hole parts are rejoined and the result is merged. The straight skeleton is useful shape descriptor of digital object. This algorithm stated here has applications in shape analysis. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Combinatorial optimization

来源：评论

学校读者我要写书评

暂无评论

Exploiting Multi-Level Data Uncertainty for Japanese-Chinese Neural Machine Translation

引用

IEICE Transactions on Information and Systems 2025年第5期E108.D卷 440-443页

作者： Li, Zezhong Ma, Jianjun Ren, Fuji Faculty of Computer Science Zhejiang University of Water Resources and Electric Power China Faculty of Foreign Languages Dalian University of Technology China College of Computer Science and Engineering The Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China China

The performance of Neural Machine Translation (NMT) heavily depends on the severity of data uncertainty existing in the training examples. In terms of its causes, data uncertainty can be categorized into intrinsic and extrinsic uncertainty, both of which can increase the learning difficulty in NMT, and lead to degradation in translation performance. To cope with this challenge, we propose a simple yet effective method to estimate the data uncertainty and incorporate it into the adaptive training of NMT, which can mitigate the hurt brought by the uncertain data. Our method consists of two modules: 1) we propose a mBERT based model for estimating token-level and sentence-level data uncertainties jointly, which is trained with multi-task learning;2) we propose heterogeneous ways of incorporating the derived multi-level uncertainties into the NMT training by using soft-masked embedding and weighted loss. Extensive experiments on Japanese↔Chinese translation show that our proposed methods substantially outperform the strong baselines in terms of BLEU scores, and verify the effectiveness of modeling data uncertainty in NMT. Copyright © 2025 The Institute of Electronics, Information and Communication Engineers.

关键词： Neural machine translation

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：