检索结果-内蒙古大学图书馆

arXiv 2020年

作者： Zhang, Wenchao Fu, Chong Xie, Haoyu Zhu, Mai Tie, Ming Chen, Junxin School of Computer Science and Engineering Northeastern University Shenyang110819 China Engineering Research Center of Security Technology of Complex Network System Ministry of Education China Key Laboratory of Intelligent Computing in Medical Image Ministry of Education Northeastern University Shenyang110819 China Science and Technology on Space Physics Laboratory Beijing100076 China College of Medicine and Biological Information Engineering Northeastern University Shenyang110819 China

RoIPool/RoIAlign is an indispensable process for the typical two-stage object detection algorithm, it is used to rescale the object proposal cropped from the feature pyramid to generate a fixed size feature map. However, these cropped feature maps of local receptive fields will heavily lose global context information. To tackle this problem, we propose a novel end-to-end trainable framework, called Global Context Aware (GCA) RCNN, aiming at assisting the neural network in strengthening the spatial correlation between the background and the foreground by fusing global context information. The core component of our GCA framework is a context aware mechanism, in which both global feature pyramid and attention strategies are used for feature extraction and feature refinement, respectively. Specifically, we leverage the dense connection to improve the information flow of the global context at different stages in the top-down process of FPN, and further use the attention mechanism to refine the global context at each level in the feature pyramid. In the end, we also present a lightweight version of our method, which only slightly increases model complexity and computational burden. Experimental results on COCO benchmark dataset demonstrate the significant advantages of our approach. Copyright © 2020, The Authors. All rights reserved.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

TCPModel: A short-term traffic congestion prediction model based on deep learning 2nd

TCPModel: A short-term traffic congestion prediction model b...

引用

2nd CCF International Conference on Artificial Intelligence, CCF-ICAI 2019

作者： Xu, Xiujuan Gao, Xiaobo Xu, Zhenzhen Zhao, Xiaowei Pang, Wei Zhou, Hongmei School of Software Dalian University of Technology Dalian116620 China Key Laboratory for Ubiquitous Network and Service Software of Liaoning Province Dalian116620 China School of Natural and Computing Sciences University of Aberdeen AberdeenAB24 3UE United Kingdom School of Transportation and Logistics Dalian University of Technology Dalian116024 China

ISBN: (纸本)9789813292970

With the progress of the urbanization, a series of traffic problems have occurred because of the growing urban population and the far lower growth rate of roads than that of cars. One of the most prominent problems is traffic congestion problem. The prediction of traffic congestion is the key to alleviate traffic congestion. To ensure the real-time performance and accuracy of the traffic congestion prediction, we propose a short-term traffic congestion prediction model called TCPModel based on deep learning. By processing a massive amount of urban taxi transportation data, we extract the traffic volume and average speed of taxis which are the most important parameters for assessing of traffic flow prediction. After analyzing the temporal and spatial distribution characteristics of the traffic flow and average speed, we present a short-term traffic volume prediction model called TVPModel, and a short-term traffic speed prediction model called TSPModel. Both models are based on a deep learning method Stacked Auto Encoder (SAE). By comparing the other traffic flow forecasting methods and average speed forecasting methods, the methods proposed by this paper have improved the accuracy rate. For traffic congestion recognition, we use a novel model called TCPModel based on three traffic parameters (average speed, traffic flow and density), which uses standard function method to standardize the parameters and calculate the congestion comprehensive threshold to determine the congestion level by thresholds. According to the experiments, TVPModel and TSPModel in this paper got satisfied accuracy compared with other prediction models. © 2019, Springer Nature Singapore Pte Ltd.

关键词： Traffic congestion

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis for Drone-Assisted HetNets with Flexible Cell Association

Performance Analysis for Drone-Assisted HetNets with Flexibl...

引用

IEEE International Conference on Communications (ICC)

作者： Hongguang Sun Xijun Wang Chao Xu Yan Zhang Tony Q.S. Quek School of Information Engineering Northwest A&F University Yangling Shaanxi China School of Electronics and Communication Engineering Sun Yat-sen University Guangzhou Guangdong China State Key Laboratory of Integrated Service Networks Xidian University Xi’an Shaanxi China Information System Technology and Design Pillar Singapore University of Technology and Design Singapore

ISBN: (数字)9781728150895

ISBN: (纸本)9781728150901

Drone small cells (DSCs) are served as aerial base stations to complement the terrestrial cellular networks, in order to provide seamless wireless coverage and increased network capacity. In this paper, we study a drone-assisted downlink heterogeneous network (HetNet) consisting of a first tier of DSCs overlaid with a second tier of terrestrial small cells (TSCs), both of which operate on the same frequency band. By considering a flexible biased association policy, we develop an analytical framework to evaluate the network performance. After deriving the association probabilities, and the probability distribution functions (PDFs) of typical link length, we derive exact expressions for the per-tier and overall coverage probabilities. The proposed framework allows to quantify the impact on network performance of most important system parameters, such as the height of DSCs, the bias factor, and the base station density. In absence of interference management, our results show that very limited gains can be obtained in the dense network scenario and the improper deployment of DSCs can only degrade the coverage probability achieved by the single-tier TSC network. What's more, the unbiased cell association is shown to be optimal for the overall coverage probability in the interference-limited network regime.

关键词： Base stations Rayleigh channels Cellular networks Downlink Heterogeneous networks Propagation losses

来源：评论

学校读者我要写书评

暂无评论

Ready for emerging threats to recommender systems? A graph convolution-based generative shilling attack

arXiv

引用

arXiv 2021年

作者： Wu, Fan Gao, Min Yu, Junliang Wang, Zongwei Liu, Kecheng Wang, Xu Key Laboratory of Dependable Service Computing in Cyber Physical Society Chongqing University Ministry of Education Chongqing401331 China School of Big Data and Software Engineering Chongqing University Chongqing401331 China School of Information Technology and Electrical Engineering The University of Queensland Queensland4072 Australia Informatics Research Centre Henley Business School University of Reading Berkshire ReadingRG6 6UD United Kingdom School of Mechanical Engineering Chongqing University Chongqing400044 China

To explore the robustness of recommender systems, researchers have proposed various shilling attack models and analyzed their adverse effects. Primitive attacks are highly feasible but less effective due to simplistic handcrafted rules, while upgraded attacks are more powerful but costly and difficult to deploy because they require more knowledge from recommendations. In this paper, we explore a novel shilling attack called Graph cOnvolution-based generative shilling ATtack (GOAT) to balance the attacks' feasibility and effectiveness. GOAT adopts the primitive attacks' paradigm that assigns items for fake users by sampling and the upgraded attacks' paradigm that generates fake ratings by a deep learning-based model. It deploys a generative adversarial network (GAN) that learns the real rating distribution to generate fake ratings. Additionally, the generator combines a tailored graph convolution structure that leverages the correlations between co-rated items to smoothen the fake ratings and enhance their authenticity. The extensive experiments on two public datasets evaluate GOAT's performance from multiple perspectives. Our study of the GOAT demonstrates technical feasibility for building a more powerful and intelligent attack model with a much-reduced cost, enables analysis the threat of such an attack and guides for investigating necessary prevention measures. Copyright © 2021, The Authors. All rights reserved.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

A Web-Based Theme-Related Word Set Construction Algorithm

引用

Asia Pacific Web (APWeb) and Web-Age Information Management (WAIM) Joint Conference on Web and Big Data, APWeb-WAIM 2018

作者： Wu, Yingkai Li, Yukun Hao, Gang Tianjin University of Technology Tianjin300384 China Tianjin Key Laboratory of Intelligence Computing and Novel Software Technology Tianjin China Key Laboratory of Computer Vision and System Ministry of Education Tianjin Tianjin China

ISBN: (纸本)9783030012977

Constructing theme-related word set is a basic work for establishing theme-oriented information retrieval systems. Nowadays, most of previous studies focus on identifying representative words of a specific document, and few studies pay attention to constructing a word set related to a theme. By analyzing existing keywords extraction methods, this paper proposes a method to automatically construct theme-related word set based on the primary theme-related word set given by domain experts and the well-known websites related to the theme. As the first step, the method uses existing information extraction techniques to obtain the documents from the websites and every document’s keyword set. Then it calculates the correlation degree between the known theme-related word set and the document keyword set, further gets a word set of the document related to the theme based on the document-theme relevance, and merges the word set to the theme-related word set. By using the method, the theme-related word set is supplemented by iteration based on the documents gotten from the theme-related websites. Because there is little research work focusing on this problem and no relevant experimental data set, this paper uses the proposed method to construct theme-related word sets towards two themes "electricity" and "college entrance examination", and we invite domain experts to evaluate the word sets. The results show that a relatively complete theme-related word set can be obtained based on this method, which shows the feasibility of our methods. © 2018, Springer Nature Switzerland AG.

关键词： Websites

来源：评论

学校读者我要写书评

暂无评论

Computation Offloading and Resource Allocation for MEC in C-RAN: A Deep Reinforcement Learning Approach 19

Computation Offloading and Resource Allocation for MEC in C-...

引用

19th IEEE International Conference on Communication technology, ICCT 2019

作者： Jin, Xiaoyan Zhang, Jun Sun, Xinghua Zhang, Ping Cai, Shu Jiangsu Key Laboratory of Wireless Communications Nanjing University of Posts and Telecommunications Nanjing210003 China Engineering Research Center of Health Service System Based on Ubiquitous Wireless Networks Ministry of Education Nanjing University of Posts and Telecommunications Nanjing210003 China School of Electronics and Communication Engineering Sun Yat-sen University Guangzhou510006 China Key Laboratory of Wireless Sensor Network and Communication Shanghai Institute of Microsystem and Information Technology Chinese Academy of Sciences Shanghai200050 China Key Laboratory of Computer Application Technology Anhui Polytechnic University Wuhu241000 China

ISBN: (纸本)9781728105352

Mobile edge computing (MEC) technology has become a promising example for cloud radio access networks (CRAN) to provide close-range services, thereby reducing service delays and saving energy consumption. In this paper, we consider a multi-user MEC system and solve the problem of the computation offloading strategies and resource allocation policies. We set the total cost of delays and energy consumption as our optimization goal. However, getting an optimal strategy in a dynamic environment is challenging. Reinforcement learning (RL) aims at long-term cumulative rewards, which are essential for time-varying dynamic systems. Therefore, we propose an optimization framework based on deep RL to solve these problems. The deep neural network (DNN) is used to estimate the value function of the critics, thereby reducing the state space complexity of the optimization target. The actor part uses another DNN to represent a parametritis stochastic strategy and improve the strategy with the help of critics. Compared with other schemes, the simulation results show that the scheme significantly reduces the total cost. © 2019 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Classifying Personal Photo Collections: An Event-Based Approach

引用

Asia Pacific Web (APWeb) and Web-Age Information Management (WAIM) Joint Conference on Web and Big Data, APWeb-WAIM 2018

作者： Geng, Ming Li, Yukun Liu, Fenglian Tianjin University of Technology Tianjin300384 China Tianjin Key Laboratory of Intelligence Computing and Novel Software Technology Tianjin China Key Laboratory of Computer Vision and System Ministry of Education Tianjin Tianjin China

ISBN: (纸本)9783030012977

Due to the increasing number of photos taken by mobile phone, how to improve the efficiency of re-finding personal photos becomes an important research issue. Based on the survey of the current methods to re-find expected photos in mobile phones, this paper proposes a photo classification strategy based on specific events. We firstly discover the relationship of shooting time and shooting location to specific events, and propose a method to classify personal photos based on them. Because there is no existing data set for experiment, we collect several representative photo collections of different people as experimental data sets, and the experiments show the efficiency of our method. We implement an album system based on our method and compare it with other existing classification ones, and the results show that our strategy can improve the efficiency of re-finding photos. © 2018, Springer Nature Switzerland AG.

关键词： Cellular telephones

来源：评论

学校读者我要写书评

暂无评论

Detection of Huanglongbing(citrus greening) based on hyperspectral image analysis and PCR

引用

Frontiers of Agricultural Science and Engineering 2019年第2期6卷 172-180页

作者： Kejian WANG Dongmei GUO Yao ZHANG Lie DENG Rangjin XIE Qiang LV Shilai YI Yongqiang ZHENG Yanyan MA Shaolan HE Southwest University/Citrus Research Institute of Chinese Academy of Agricultural Sciences Chongqing 400712 China National Agricultural Technology Extension and Service Center Beijing 100125 China Chengdu Plant Quarantine Station Chengdu 610000 China Key Laboratory of Modern Precision Agriculture System Integration Research China Agricultural University Beijing 100083 China

Huanglongbing (HLB, citrus greening) is one of the most serious quarantine diseases of citrus worldwide. To monitor in real-time, recognize diseased trees, and efficiently prevent and control HLB disease in citrus, it is necessary to develop a rapid diagnostic method to detect HLB infected plants without symptoms. This study used Newhall navel orange plants as the research subject, and collected normal color leaf samples and chlorotic leaf samples from a healthy orchard and an HLB-infected orchard, respectively. First, hyperspectral data of the upper and lower leaf surfaces were obtained, and then the polymerase chain reaction (PCR) was used to detect the HLB bacterium in each leaf. The PCR test results showed that all samples from the healthy orchard were negative, and a portion of the samples from the infected orchard were positive. According to these results, the leaf samples from the orchards were divided into disease-free leaves and HLB-positive leaves, and the least squares support vector machine recognition model was established based on the leaf hyperspectral reflectance. The effect on the model of the spectra obtained from the upper and lower leaf surfaces was investigated and different pretreatment methods were compared and analyzed. It was observed that the HLB recognition rate values of the calibration and validation sets based on upper leaf surface spectra under 9-point smoothing pretreatment were 100% and 92.5%, respectively. The recognition rate values based on lower leaf surface spectra under the second-order derivative pretreatment were also 100% and 92.5%, respectively. Both upper and lower leaf surface spectra were available for recognition of HLB-infected leaves, and the HLB PCR-positive leaves could be distinguished from the healthy by the hyperspectral modeling analysis. The results of this study show that early and nondestructive detection of HLBinfected leaves without symptoms is possible, which provides a basis for the hyperspectral diagnosi

关键词： citrus HLB hyperspectral identification PCR

来源：评论

学校读者我要写书评

暂无评论

CSF: An Efficient Parallel Deduplication Algorithm by Clustering Scattered Fingerprints

CSF: An Efficient Parallel Deduplication Algorithm by Cluste...

引用

IEEE International Conference on Big Data and Cloud computing (BdCloud)

作者： Hao Fan Guangping Xu Yi Zhang Liming Yuan Yanbing Xue Tianjin University of Technology Tianjin China School of Computer Science and Engineering Tianjin University of Technology Tianjin China Key Laboratory of Computer Vision and System Ministry of Education Tianjin China Tianjin Key Laboratory of Intelligence Computing and Novel Software Technology Tianjin China

ISBN: (数字)9781728143286

ISBN: (纸本)9781728143293

Deduplication is one of the most effective and efficient techniques to save memory space. It is widely used in data centers and cloud storage systems. Multi-stream concurrency is expected to increase the throughput of deduplication. However, multiple data streams hurt the locality of accessed data and weaken the benefit of data concurrency, which forms a challenge for data deduplication. Usually, the ordered index can reshape the locality of data streams, which can improve the cache hit rate during deduplication. In this paper, we first propose an efficient parallel deduplication algorithm by clustering scattered fingerprints, called CSF, to exploit the data locality as much as possible. It tries to improve the utilization rate of the fingerprint page by the clustered fingerprints. Moreover, it retains the scattered fingerprint to next round fingerprint comparison by re-using the fingerprints on the same page. Thus the number of the fingerprint pages to read is reduced. We further optimize the proposed algorithm by a scheduling strategy, which effectively schedules the task of part streams ahead while ensuring the overall performance. Finally, we evaluated the performance of our algorithm with various data sets in experiments. The experimental results show that our proposed algorithm achieves better performance than the state-of-the-art method.

关键词： Indexes Clustering algorithms Task analysis Servers Fingerprint recognition Throughput Distributed databases

来源：评论

学校读者我要写书评

暂无评论

Preventing Spread of Spam Transactions in Blockchain by Reputation

Preventing Spread of Spam Transactions in Blockchain by Repu...

引用

International Workshop on Quality of service

作者： Jiarui Zhang Yukun Cheng Xiaotie Deng Bo Wang Jan Xie Yuanyuan Yang Mengqian Zhang Stony Brook University Stony Brook NY USA School of Business Suzhou Key Laboratory for Big Data and Information Service Suzhou University of Science and Technology Suzhou China Centers on Frontiers of Computing Studies Peking University Beijing China Hangzhou Cryptape Technology Co. Ltd. Hangzhou China Shanghai Jiao Tong University Shanghai China

ISBN: (数字)9781728168876

ISBN: (纸本)9781728168883

As one of the fastest-growing applications in the Peer-to-Peer (P2P) network, the development of blockchain technology is accompanied by different attacks. Those include whitewashing, free-riding, and distributed denial of service (DDoS) attacks, particularly because of features such as anonymity, distributed, permissionless in the blockchain network. One popular of them is spam transactions. Although the blockchain protocol requires each node to verify all received transactions, many nodes choose to forward transactions without verification to conserve their computational power, as there is no punishment for such a shirking. And it makes the blockchain vulnerable to the spreading of spam transactions over the network and creates extra burdens for all nodes in the network. We propose a reputation mechanism for the blockchain system to tackle this problem: Each node will locally compute reputations of its neighbors, and decide the probability to verify a received transaction based on the reputation value of the transaction sender. In turn, its neighbors will have an incentive to conduct verification to keep its reputation high. Subsequently, spam transactions can be blocked before reaching the miners. We have conducted a series of simulations, which clearly demonstrate the advantage of our reputation mechanism.

关键词： Peer-to-peer computing Computer crime Bitcoin Quality of service Earth Observing system

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：