检索结果-内蒙古大学图书馆

Energy-efficient computation offloading via deep reinforcement learning in mobility-aware multi-access edge computing systems with diverse users

引用

Expert Systems with Applications 2025年 286卷

作者： Wu, Haixing Jin, Shunfu School of Information Science and Engineering Yanshan University Qinhuangdao066004 China Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Yanshan University Qinhuangdao066004 China

Multi-access Edge Computing (MEC) has been a promising solution that enables Internet of Things (IoT) devices to support computation-intensive applications by offloading some tasks to the network edge. However, most existing offloading methods in MEC systems are based on the premise of scenarios with either single-type or stable users, which is not aligned with the practical situations of users’ diversity and mobility. In addition, these methods always ignore energy saving strategies in MEC systems, which inevitably degrades Utility Energy Efficiency (UEE). To tackle these issues, considering personalized requirements from diverse users and idle energy consumption in MEC servers, we propose a novel computation offloading architecture with adaptive sleeping mechanism and heterogeneous MEC servers. Based on the architecture, we investigate energy-efficient computation offloading in mobility-aware MEC systems with diverse users. Under the Deep Reinforcement Learning (DRL) framework, we propose a Twin Delayed Deep Deterministic Policy Gradient Algorithm with Bursty Traffic (TD3-BT), in which the inherent correlation of task arrivals is taken into account to capture the behavior of correlated traffic and provide rational metrics estimation within one time slot. A long-term system cost minimization problem is formulated to optimize computation offloading for the trade-off between system delay and UEE. Assisted by TD3-BT, the formulated problem is effectively solved and the decision making in each time slot is provided for diverse users. Experimental results demonstrate that our TD3-BT algorithm achieves superior performance under various offloading scenarios compared to some benchmarks. © 2025 Elsevier Ltd

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

An Ensemble Detection Method for Shilling Attacks Based on Features of Automatic Extraction

引用

China Communications 2019年第8期16卷 130-146页

作者： Yaojun Hao Fuzhi Zhang Jinbo Chao School of Information Science and Engineering Yanshan UniversityQinhuangdao 066004China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao 066004China Department of Computer Science and Technology Xinzhou Teachers UniversityXinzhou 034000China

Faced with the evolving attacks in recommender systems, many detection features have been proposed by human engineering and used in supervised or unsupervised detection methods. However, the detection features extracted by human engineering are usually aimed at some specific types of attacks. To further detect other new types of attacks, the traditional methods have to re-extract detection features with high knowledge cost. To address these limitations, the method for automatic extraction of robust features is proposed and then an Adaboost-based detection method is presented. Firstly, to obtain robust representation with prior knowledge, unlike uniform corruption rate in traditional mLDA(marginalized Linear Denoising Autoencoder), different corruption rates for items are calculated according to the ratings’ distribution. Secondly, the ratings sparsity is used to weight the mapping matrix to extract low-dimensional representation. Moreover, the uniform corruption rate is also set to the next layer in mSLDA(marginalized Stacked Linear Denoising Autoencoder) to extract the stable and robust user features. Finally, under the robust feature space, an Adaboost-based detection method is proposed to alleviate the imbalanced classification problem. Experimental results on the Netflix and Amazon review datasets indicate that the proposed method can effectively detect various attacks.

关键词： shilling attack ensemble detection features of automatic extraction marginalized linear denoising autoencoder

来源：评论

学校读者我要写书评

暂无评论

Research on a Novel Influence Maximization Algorithm Based on Community Structure

Research on a Novel Influence Maximization Algorithm Based o...

引用

作者： Jing Chen Jiangchuan Liu College of Information Science and Engineering Yanshan University Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Software Engineering Laboratory in Hebei Province

With the research of influence maximization algorithm, many researchers have found that the existing algorithm has the problem of overlapping influence of seed nodes. In order to solve the problem of overlapping influence of seed nodes, this paper proposes an IMCS algorithm based on community structure. Firstly, we divide the community through the central node, and the quality of community division is ensured by defining community fitness and node contribution. Then through the analysis of the community division results, the seed node selects the one with the largest degree. Since most of the nodes activated by seed nodes of different communities also belong to different communities, this method solves the problem of overlapping influence to a certain extent. The experimental results show that the effectiveness of the IMCS algorithm is verified under the real network, cooperative network and artificial network, and the IMCS algorithm has a better effect than IEIR and Degree algorithms in most networks under the IC model.

关键词： Seed node influence maximization community division overlapping influence community structure independent cascade model

来源：评论

学校读者我要写书评

暂无评论

Research on the Influence of Improved K-shell Algorithm on Commodity Profit

Research on the Influence of Improved K-shell Algorithm on C...

引用

作者： Jing Chen Jincheng Huang College of Information Science and Engineering Yanshan University Key Laboratory for Computer Virtual Technology and system Integration of Hebei Province Software Engineering Laboratory in Hebei Province

In order to maximize the influence of commodity profits in e-commerce platforms, designing and improving the K-shell algorithm to select the more influential seed node sets in this paper. The new algorithm improves the number of active nodes by setting node threshold and edge weight attributes. To obtain more commodity profits, a strategy IRDSN(Strategy for Improving Repeat Degree of Seed Nodes) is proposed to select initial seed nodes and improve the repeat degree of seed nodes. The profit maximization based on linear threshold model is realized by setting different propagation modes. The improved algorithm and strategy IRDSN are analysed and verified in real data set and e-commerce platform. The results show that the algorithm effectively improves the profit of commodities.

关键词： Influence threshold K-shell influence maximization profit

来源：评论

学校读者我要写书评

暂无评论

Communication-Based Attacks Detection in Android Applications

引用

Tsinghua Science and technology 2019年第5期24卷 596-614页

作者： Chuan Ma Tao Wang Limin Shen Dongkui Liang Shuping Chen Dianlong You School of In formation Science and Engineering Yanshan University the Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao 066004 China School of Business Ad mi nisliatiom Hebei Normal University of Science and Technology. Qinhuangdao 066004 China Library of Yanshan University. Yanshan University Qinhuangdao 066004 China

The Android operating system provides a rich Inter-Component Communication(ICC) method that brings enormous convenience. However, the Android ICC also increases security risks. To address this problem, a formal method is proposed to model and detect inter-component communication behavior in Android applications. Firstly,we generate data flow graphs and data facts for each component through component-level data flow ***, our approach treats ICC just like method calls. After analyzing the fields and data dependencies of the intent, we identify the ICC caller and callee, track the data flow between them, and construct the ICC model. Thirdly,the behavior model of Android applications is constructed by a formal mapping method for component data flow graph based on Pi calculus. The runtime sensitive path trigger detection algorithm is then given. Communicationbased attacks are detected by analyzing intent abnormity. Finally, we analyze the modeling and detection efficiency,and compare it with relevant methods. Analysis of 57 real-world applications partly verifies the effectiveness of the proposed method.

关键词： Android inter-component communication intents component hijacking attack detection

来源：评论

学校读者我要写书评

暂无评论

A PSO Based Multi-Domain virtual Network Embedding Approach

引用

China Communications 2019年第4期16卷 105-119页

作者： Yongjing Ni Guoyan Huang Sheng Wu Chenxi Li Peiying Zhang Haipeng Yao College of Information Science and Engineering Yanshan University College of Information Science and Engineering Hebei University of Science and Technology The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Yanshan University School of Information and Communication Engineering Beijing University of Posts and Telecommunications College of Computer & Communication Engineering China University of Petroleum (East China)

This paper proposed a multi-domain virtual network embedding algorithm based on multi-controller SDN architecture. The local controller first selects candidate substrate nodes for each virtual node in the domain. Then the global controller abstracts substrate network topology based on the candidate nodes and boundary nodes of each domain, and applies Particle Swarm Optimization Algorithm on it to divide virtual network requests. Each local controller then embeds the virtual nodes of the divided single-domain virtual network requests in the domain, and cooperates with other local controllers to embed the inter-domain virtual links. Simulation experimental results show that the proposed algorithm has good performance in reducing embedding cost with good stability and scalability.

关键词： multi-domain virtual network embedding candidate node particle swarm optimization algorithm virtual network request division

来源：评论

学校读者我要写书评

暂无评论

Multi-Update Patterns and Validity Verification for Robust Visual Tracking

引用

Journal of computers (Taiwan) 2020年第4期31卷 77-90页

作者： Wen, Jia Wang, Hong-Jun Li, Yan-Nan Zhao, Ji-Wei Yang, Yu-Ying School of Information Science and Engineering Yanshan University Qinhuangdao China Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province China Key Laboratory of Software Engineering of Hebei Province China

The Object tracking is a challenging problem in computer vision field. Now, deep learning has made outstanding achievements in feature extraction. There are already some examples of deep learning applications in visual tracking. But, they almost paid attention to precision and sacrificed efficiency. So we propose a tracker named adaptive combination tracker(ACT), based on neural network model, which gives consideration to both precision and efficiency. Our ACT tracker gets rich features offline through Stacked Denoising Autoencoders (SDAE) and transfers them to online tracking. In tracking, a dynamic particle filter algorithm is proposed to speed up tracking. We update our tracker with different patterns according to the change in appearance. In order to ensure efficiency, our tracker regulates speed adaptively. Our tracker regulates speed precisely to avoid drift caused by fast motion as much as possible, and confirm and optimize the effectiveness of the results by comparison with previous target. We evaluate our tracker on OTB2013 datasets. Extensive experiments on OTB2013 datasets demonstrate that the proposed tracker performs favorably in relation to the state-of-the-art methods. © 2020 computer Society of the Republic of China. All rights reserved.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

SIGAN: A novel image generation method for solar cell defect segmentation and augmentation

arXiv

引用

arXiv 2021年

作者： Su, Binyi Zhou, Zhong Chen, Haiyong Cao, Xiaochun State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing100191 China School of Artificial Intelligence and Data Science Hebei University of Technology Tianjin300130 China State Key Laboratory of Information Security Institute of Information Engineering Chinese Academy of Sciences Beijing100093 China Peng Cheng Laboratory Shenzhen518055 China

Solar cell electroluminescence (EL) defect segmentation is an interesting and challenging topic. Many methods have been proposed for EL defect detection, but these methods are still unsatisfactory due to the diversity of the defect and background. In this paper, we provide a new idea of using generative adversarial network (GAN) for defect segmentation. Firstly, the GAN-based method removes the defect region in the input defective image to get a defect-free image, while keeping the background almost unchanged. Then, the subtracted image is obtained by making difference between the defective input image with the generated defect-free image. Finally, the defect region can be segmented through thresholding the subtracted image. To keep the background unchanged before and after image generation, we propose a novel strong identity GAN (SIGAN), which adopts a novel strong identity loss to constraint the background consistency. The SIGAN can be used not only for defect segmentation, but also small-samples defective dataset augmentation. Moreover, we release a new solar cell EL image dataset named as EL-2019, which includes three types of images: crack, finger interruption and defect-free. Experiments on EL-2019 dataset show that the proposed method achieves 90.34% F-score, which outperforms many state-of-the-art methods in terms of solar cell defects segmentation results. Copyright © 2021, The Authors. All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Local outlier detection algorithm based on gaussian kernel density function 10th

Local outlier detection algorithm based on gaussian kernel d...

引用

10th International Symposium on Intelligence Computation and Applications, ISICA 2018

作者： Zhang, Zhongping Liu, Jiaojiao Miao, Chuangye School of Information Science and Engineering Yanshan University QinhuangdaoHebei066004 China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province QinhuangdaoHebei066004 China

ISBN: (纸本)9789811364723

With the rapid development of information technology, the structure of data resources is becoming more and more complex, and outlier mining is attracting more and more attention. Based on Gaussian kernel function, this paper considers three neighbors: k nearest neighbors, reverse k neighbors and shared nearest neighbors. A local outlier detection algorithm based on Gaussian kernel function is proposed. Firstly, the algorithm stores the nearest neighbors of each data object through kNN maps, including k-nearest neighbors, reverse k-nearest neighbors, and shared nearest neighbors, forming a kernel neighbor set S. Secondly, Estimating density of data objects through kernel density estimation KDE method. Finally, the relative density outlier factor RDOF is used to estimate the degree of data objects deviating from the neighborhood, and then determines whether the data objects are outliers, and the validity of the algorithm is proved on the real and synthetic data sets. © Springer Nature Singapore Pte Ltd. 2019.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Mining important nodes in complex software network based on ripple effects of probability 19

Mining important nodes in complex software network based on ...

引用

2019 ACM Turing Celebration Conference - China, ACM TURC 2019

作者： Ren, Jiadong Huang, Guoyan Wang, Qian He, Haitao Liu, Xinqian Zhao, Xiaolin Computer Virtual Technology and System Integration Laboratory of Hebei Province College of Information Science and Engineering Yanshan University Qinhuangdao Hebei China Beijing Key Laboratory of Software Security Engineering Technology School of Computer Science and Technology Beijing Institute of Technology Beijing China

ISBN: (纸本)9781450371582

The complexity of software directly leads to an increasing cost in software testing and maintenance. Finding the important nodes with significant vulnerability is helpful for fault discovery and further reduces the damage to the software system. In this paper, a new algorithm named MIN-REP (Mining the Important Nodes based on Ripple Effects of Probability) is proposed to find out the paths with greater possibility for fault propagation, and then the important nodes are mined. To build a model of directed unweighted software network, functions are taken as the nodes and the dependencies between the functions are regarded as the edges. Fault propagation tendency paths are discovered based on the function execution paths and minimum probability threshold. The frequency of each directed edge in the set of fault propagation tendency path is taken as the weight of the corresponding edge. Then some metrics related to ripple effects of probability are calculated. Finally, the nodes with the metric at top-k are taken as the important nodes. The experiment verifies the accuracy and efficiency of the algorithm MIN-REP. © 2019 Association for Computing Machinery.

关键词： Probability

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：