检索结果-内蒙古大学图书馆

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Shihui Zhang Ziteng Xue Yuhong Jiang Houlin Wang School of Information Science and Engineering Yanshan University Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province

The Amodal Instance Segmentation (AIS) task aims to infer the visible and occluded regions of an object instance. Existing AIS methods typically focus on directly predicting visible and occluded regions or leveraging prior knowledge to guide predictions. However, these methods often ignore the perception of occluded views, leading to inaccurate results. To address this issue and achieve high-quality AIS, we propose a boundary-aware Occlusion Perception Network (OPNet). OPNet consists of three main components: the Dynamic Feature Augmentation Pyramid (DFAP), the Dual-path Boundary Aware Module (DBAM), and the Shape-guide Refinement Module (SRM). Specifically, DBAM employs an occlusion-perception strategy to learn discriminative features with boundary information, enabling it to distinguish occlusion from multiple views. Additionally, DFAP and SRM optimize the results by enhancing feature aggregation and imposing geometric constraints. Experiments on the D2SA, KINS, and CWALT datasets show that OPNet significantly outperforms state-of-the-art AIS methods that without prior knowledge. Code is available at https://***/ZitengXue/OPNet.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Robust Recommendation Method Based on Shilling Attack Detection and Matrix Factorization Model

Robust Recommendation Method Based on Shilling Attack Detect...

引用

2017 2nd International Conference on Communications, Information Management and Network Security (CIMNS2017)

作者： Yu-qi HU Kai LIU Fu-zhi ZHANG School of Information Science and Engineering Yanshan University The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province

The existing robust collaborative recommendation algorithms have low robustness against PIA and Ao P attacks. Aiming at the problem, we propose a robust recommendation method based on shilling attack detection and matrix factorization model. Firstly, the type of shilling attack is identified based on statistical characteristics of attack profiles. Secondly, we devise corresponding unsupervised detection algorithms for standard attack, Ao P and PIA, and the suspicious users and items are flagged. Finally, we devise a robust recommendation algorithm by combining the proposed shilling attack detection algorithm with matrix factorization model, and conduct experiments on the Movie Lens dataset to demonstrate its effectiveness. Experimental results show that the proposed method exhibits good recommendation precision and excellent robustness for shilling attacks of multiple types.

关键词： Collaborative recommendation Shilling attack Attack type identification Attack detection Matrix factorization model

来源：评论

学校读者我要写书评

暂无评论

Research on Router Algorithm Based on Trust Primary Node in the Environment of P2P Network

Research on Router Algorithm Based on Trust Primary Node in ...

引用

2015 International Conference on Informatics, Control and Automation(ICA 2015)

作者： Jing CHEN Yun WAN Department of Computer College of Information Science and EngineeringYanshan University Key Laboratory of Computer Virtual Technology and System Integration of Hebei Province

The distributed, heterogeneous and anonymous attribute of P2 P networks make the node information sharing facing security problems. In order to prevent malicious and false service of nodes, and improve the retrieval efficiency, the primary node selection algorithm based on trust mechanism is proposed in this paper. Firstly, the network topology based on virtual community is described, and the function of primary node is discussed. Secondly, the primary node is selected according to direct trust value and recommendation trust value of each node in the virtual community. Finally, resource retrieval algorithm is designed according to the selected trust primary node, and the algorithm is experimentally verified. In some certain extent, the results show that the method is able to reduce the fraud of primary node and improve the retrieval efficiency.

关键词： P2P network virtual community Primary node Trust mechanism Router algorithm

来源：评论

学校读者我要写书评

暂无评论

Research on community detection in social network based on Pairwise Merging

引用

Journal of Computational Information systems 2015年第19期11卷 7079-7086页

作者： Chen, Jing Wan, Yun Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao China College of Information Science and Engineering Yanshan University Qinhuangdao China

Community detection has become an important challenge during the past decade. Network is divided into some groups or communities that are densely connected to each other inside while less connected to the nodes outside. Community detection in social networks using hybrid merging of sub-communities is proposed. It is based on the idea of bottom to up. But the initial phase of algorithm resulted in many sub- communities and hybrid merging generated some weak communities. In order to resolve these problems, the algorithm CDPM (Community Detection based on Pairwise Merging of sub-communities) is proposed in this paper. The main idea of the algorithm starts with some small communities. Then, these small communities are extended by using similarity in order to decrease the number of sub-communities. Finally, these sub-communities are merged by pairwise merging which is mainly based on maximization modularity. Meanwhile, checking structure of every community is to avoid the appearance of weak community. The CDPM is proved on two real networks, and the results show that the performance of CDPM is close to real network and the quality of community detection is improved. © 2015 by Binary Information Press.

关键词： Merging

来源：评论

学校读者我要写书评

暂无评论

Community detection algorithm in signed social networks based on statistics and merger

引用

Journal of Information and Computational Science 2015年第15期12卷 5589-5599页

作者： Guo, Jingfeng Zhao, Yue Hu, Xinzhuan Liu, Yuanying College of Information Science and Engineering Yanshan University Qinhuangdao China Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao China

Researches on community detection in signed social networks focus on the assignment of positive and negative edges. However, the community detection approaches that positive and negative edges are handled separately in two phases cannot make full use of negative edges, which seriously affects the accuracy of community detection. To solve the problem, a new algorithm called CDA SM (Community Detection Algorithm Based on Statistics and Merger) is proposed: merge into the two phases as an unified process;in the process of community detection, the node shall be added to the community in which the node has the maximum contribution as well as the minimum interference degree. The accuracy and time efficiency of CDA SM have been demonstrated through a set of experiments including both un-weighted common networks and un-weighted signed social networks. © 2015 by Binary Information Press

关键词： Signal detection

来源：评论

学校读者我要写书评

暂无评论

引用

Journal of Computational Information systems 2015年第2期11卷 535-545页

作者： Ren, Jiadong Li, Weina Han, Gaowei College of Information Science and Engineering Yanshan University Qinhuangdao China Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao China

Dissimilar path detection is an important task in software behavior detection. Previous dissimilar path mining algorithms ignore time-interval weight and dissimilarity. Therefore, a dissimilar path detecting algorithm called Graph-Dissimilar-Detect in complex software behavior network based on time-interval weight and dissimilarity is put forward. Firstly, weighted complex software behavior network model is built. DFS is adopted for the network to obtain paths. Secondly, Path-Length-Constraint is designed to prune some useless paths. Time-interval-Constraint is devised to prune some weak relation between software behaviors. Strong-Path-Checking is developed to prune weighted key sub-paths. Thirdly, KPath-Mine is designed to mine weighted key paths. Fourthly, definitions of dissimilar paths and path dissimilarity are provided and Dissimilar-Path-Mine is devised to find dissimilar paths. Experimental results show Graph-Dissimilar-Detect algorithm has a good scalability in detecting dissimilar paths in complex software behavior network. Furthermore, considering time-interval and path dissimilarity, more interesting and accurate key paths and dissimilar paths can be found. 1553-9105/Copyright © 2015 Binary Information Press

关键词： Graph theory

来源：评论

学校读者我要写书评

暂无评论

Analysis on impact of node failure in software execution network

引用

Journal of Computational Information systems 2015年第6期11卷 2217-2225页

作者： He, Haitao Ren, Rong Zhang, Bing Ren, Jiadong College of Information Science and Engineering Yanshan University Qinhuangdao China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao China

One or few failure nodes will cause cascading failure in software network, which would bring about security issues. Thus, analyzing nodes becomes an indispensable aid in software security. In this paper, software execution process is abstracted as weighted complex network. Based on this, a cascading failure diffusion analyzing (CFDA) algorithm is put forward to evaluate the impact of node failure on software during its runtime dynamically. It explores nodes' failure spreading range and the number of infected nodes by malicious attacks on high in-strength nodes and strong failure propagation probability nodes in software execution networks respectively. Meanwhile, the propagation behavior of cascading failures in the software system is studied. The experiment results prove that strong failure propagation ability nodes under attack have greater devastating impact on software than others. Copyright © 2015 Binary Information Press.

关键词： Complex networks

来源：评论

学校读者我要写书评

暂无评论

The synthesis of dynamic junction over-constrained parallel-parallel mechanism based on the weighted directed graph theory

引用

ICIC Express Letters 2013年第11期7卷 3057-3065页

作者： Guo, Xijuan Zhang, Dayu Geng, Qingjia Zhang, Wenyuan The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Yanshan University No. 438 Hebei Ave. Qinhuangdao 066004 China

Several dynamic junction over-constrained parallel-parallel mechanisms are proposed in this paper with a new method based on graph theory synthesis. This kind of mechanism is the pivotal part of the equivalent mechanism in helicopter automatic tilting device. The purpose of synthesis is that the synthesis result is helpful to optimize the structure and optimal design for helicopter automatic tilting device, and the reason why a graph method is chosen is that it can be programmed easily. The characteristic of "Dynamic Junction" is proposed relative to fixed junction. Because the joint of connecting two parallel mechanisms is dynamic, these two mechanisms can motion relatively. A mount of permutation and combination calculating processes are needed because of dynamic characteristic. In this view, choosing a synthesis method which can program easily in computer is appropriate. The graph method is the best choice. However, classic graph method is difficult to discriminate the graph isomorphism and limited to apply in spatial mechanism. Therefore, this paper proposed a new graph way. First, mechanism is modeled in weighted directed graph, then based on equal freedom of degree, many other graph models are got which have the same motion characteristic. Finally many target mechanisms can be derived by those models. The new method avoids the problem of classic graph method and is easy to program and run in computer, it is proved effectively because several dynamic junction over-constrained parallel-parallel mechanisms are successfully synthesized by it. © 2013 ISSN 1881-803X.

关键词： Dynamics

来源：评论

学校读者我要写书评

暂无评论

Grid resource dynamic replication strategy based on access frequency and time cost

引用

Journal of Computational Information systems 2015年第6期11卷 1973-1982页

作者： Zhang, Zhongping He, Long The School of Information Science and Engineering Yanshan University Qinhuangdao China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao China

Data grid provides a large amount of computing resources and storage space to handle data-intensive applications. Because of the limitation of available resources and the production of large-scale workload in grid, the effective use of grid resources has become a major challenge. Data replication can reduce the bandwidth consumption, improve the fault-tolerant ability and enhance the extensibility. A grid resources dynamic replication strategy based on the access frequency and time cost is given in this paper. The proposed strategy deletes the file by considering two important parameters: access frequency and transmission costs. In the replica selection strategy, the best replica is selected by considering the waiting time of the replica request queue in the storage and the data transfer time. The simulated results with OptorSim show that the proposed replication strategy gives better performance compared with DHRA algorithm. Copyright © 2015 Binary Information Press.

关键词： Data transfer

来源：评论

学校读者我要写书评

暂无评论

A virtualized data center energy-saving mechanism based on switching operating mode of physical servers and reserving virtual machines

A virtualized data center energy-saving mechanism based on s...

引用

作者： Yin, Chunxia Liu, Jian Jin, Shunfu College of Information Science and Engineering Yanshan University Qinhuangdao China Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Yanshan University Qinhuangdao China Ocean College of Hebei Agricultural University Qinhuangdao China Institute of Computing Technology Chinese Academy of Sciences Beijing China

The energy consumption of virtualized data centers has grown very fast in last several years. Because a large number of hosts are running in an idle state, virtualized data centers waste a large amount of electric energy. To save more energy for virtualized data centers, an energy-saving mechanism is proposed based on switching operating mode of physical servers and reserving virtual machines (VMs). The main idea is that when the amount of idle VMs reaches twice of a specified threshold, half of these idle VMs are reserved to process the new task that is about to arrive, and the physical server which hosts the other half of the idle VMs is switched to the sleep mode. From the perspective of task arrival rate and sleep parameters, we use two-dimensional Markov processes to analyze the proposed energy-saving mechanisms. By using matrix geometry solutions, we theoretically estimate energy consumption and response performance. According to the numerical experiments, the proposed energy-saving mechanism obviously cut down electric energy consumption and ensures response performance. Finally, the specified threshold for the amount of VMs is optimized by building a cost function. © 2020 John Wiley & Sons, Ltd.

关键词： Cost functions

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：