检索结果-内蒙古大学图书馆

2018 International Conference on Network, Communication, Computer engineering (NCCE 2018)

作者： Dongyang Liu Jindong Zhang Xiaofeng Zhang Xiaoyu Ma Yuxin Kou College of Computer Science and Technology Jilin University Key Laboratory of Symbol Computation and Knowledge Engineering of the Ministry of Education Jilin University College of Software Jilin University

ISBN: (纸本)9781510871076

With the continuous development of machine vision,intelligent detection of bottom spray glue has gradually become a *** this paper,we proposed an algorithm of detection the glue spray on bottom of vehicle based on machine *** sampling and detecting the spray interest area,the algorithm can obtain its features which were used for intelligent identification of the spray interest area,and then carry out the pseudo target and detect whether the detection area is complete or *** can test the vehicles on pipeline in real *** the average,each frame's processing time is less than 150 *** there is an unqualified area,it will alarm and inform the operators of the wrong *** algorithm can improve the quality of the vehicle and save human resources and identify the spray interest area with high *** research directions in the future is to accelerate the speed to meet the practical application requirements,and to solve the intelligent identification of the spray interest area which has the same color with the vehicle body.

关键词： Glue spray Machine vision intelligent detection

来源：评论

学校读者我要写书评

暂无评论

ADAPTIVE DEPTH GRAPH ATTENTION NETWORKS

arXiv

引用

arXiv 2023年

作者： Zhou, Jingbo Du, Yixuan Zhang, Ruqiong Zhang, Rui Key Laboratory of Symbolic Computation and Knowledge Engineering of MOE Jilin University China

As one of the most popular GNN architectures, the graph attention networks (GAT) is considered the most advanced learning architecture for graph representation and has been widely used in various graph mining tasks with impressive results. However, since GAT was proposed, none of the existing studies have provided systematic insight into the relationship between the performance of GAT and the number of layers, which is a critical issue in guiding model performance improvement. In this paper, we perform a systematic experimental evaluation and based on the experimental results, we find two important facts: (1) the main factor limiting the accuracy of the GAT model as the number of layers increases is the oversquashing phenomenon;(2) among the previous improvements applied to the GNN model, only the residual connection can significantly improve the GAT model performance. We combine these two important findings to provide a theoretical explanation that it is the residual connection that mitigates the loss of original feature information due to oversquashing and thus improves the deep GAT model performance. This provides empirical insights and guidelines for researchers to design the GAT variant model with appropriate depth and well performance. To demonstrate the effectiveness of our proposed guidelines, we propose a GAT variant model-ADGAT that adaptively selects the number of layers based on the sparsity of the graph, and experimentally demonstrate that the effectiveness of our model is significantly improved over the original GAT. Copyright © 2023, The Authors. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

RiWalk: Fast Structural Node Embedding via Role Identification

RiWalk: Fast Structural Node Embedding via Role Identificati...

引用

IEEE International Conference on Data Mining (ICDM)

作者： Xuewei Ma Geng Qin Zhiyang Qiu Mingxin Zheng Zhe Wang College of Computer Science and Technology Jilin University Changchun China Key Laboratory of Symbol Computation and Knowledge Engineering Ministry of Education Jilin University Changchun China

Nodes performing different functions in a network have different roles, and these roles can be gleaned from the structure of the network. Learning latent representations for the roles of nodes helps to understand the network and to transfer knowledge across networks. However, most existing structural embedding approaches suffer from high computation and space cost or rely on heuristic feature engineering. Here we propose RiWalk, a flexible paradigm for learning structural node representations as it decouples the structural embedding problem into a role identification procedure and a network embedding procedure. Through role identification, rooted kernels with structural dependencies kept are built to better integrate network embedding methods. To demonstrate the effectiveness of RiWalk, we develop two different role identification methods named RiWalk-SP and RiWalk-WL respectively and employ random walk based network embedding methods. Experiments on within-network classification tasks show that our proposed algorithms achieve comparable performance with other baselines while being an order of magnitude more efficient. Besides, we also conduct across-network role classification tasks. The results show potential of structural embeddings in transfer learning. RiWalk is also scalable, making it capable of capturing structural roles in massive networks.

关键词： Kernel Task analysis knowledge engineering Machine learning Bridges Proteins Network topology

来源：评论

学校读者我要写书评

暂无评论

Optimization strategy of top-down join enumeration on modern multi-core CPUs

引用

Journal of Computers 2011年第10期6卷 2004-2012页

作者： Zuo, Wanli Chen, Yongheng He, Fenglin Chen, Kerui Key Laboratory of Symbol Computation and Knowledge Engineering of the Ministry of Education College of Computer Science and Technology Jilin University Changchun China College of Computer Science and Technology Jilin University Changchun China

Most contemporary database systems query optimizers exploit System-R's bottom-up dynamic programming method (DP) to find the optimal query execution plan (QEP) without evaluating redundant subplans. The distinguished exceptions are Volcano/Cascades using transforms to generate new plans according to a topdown approach. As recent research has revealed, bottom-up dynamic programming can improve performance with respect to the shape of the join graph and parallelism. However top-down join enumeration dynamic programming method can derive upper bounds for the costs of the plans it generates which is not available to typical bottom-up DP method. In this paper, we propose a comprehensive and practical framework for parallelizing top-down dynamic programming query optimization with complex non-inner join in the multi-core processor architecture, referred as PTDhyp. We have implemented such a search strategy and experimental results show that can improve optimization time effective compared to known existing algorithms. © 2011 ACADEMY PUBLISHER.

关键词： Dynamic programming

来源：评论

学校读者我要写书评

暂无评论

Live Video Streaming Authentication for Vehicle Multimedia based on Merkle Tree

Live Video Streaming Authentication for Vehicle Multimedia b...

引用

2018 International Conference on Network, Communication, Computer engineering (NCCE 2018)

作者： Dongqi Han Jindong Zhang Yang Liu Peibin Wu Yiding Sun College of Computer Science and Technology Jilin University Key Laboratory of Symbol Computation and Knowledge Engineering of the Ministry of Education Jilin University State Key Laboratory of Automobile Simulation and Control Jilin University

ISBN: (纸本)9781510871076

Merkle tree has been widely applied in the field of P2 P *** this paper,an efficient authentication scheme is proposed in the integrity check module for vehicle live video streaming system by optimizing the construction and verification of the Merkle *** can create and check nodes at the same time in a shorter time without trusted *** the same time,in order to prevent the oversensitivity of hash algorithm,this paper uses the perceptual hash algorithm to generate the hash value of nodes in the Merkle *** performance evaluation and analysis,our algorithm can build a Merkle tree with 2 N-1 nodes only after the N+log2 N times hash *** has better time performance than the traditional tree,whether receiving attacks or *** it is proved that the scheme is secure under the condition that the vehicle live video streaming system is safe.

关键词： vehicle streaming media system Merkle tree P2P-streaming integrity authentication perceptual hash algorithm

来源：评论

学校读者我要写书评

暂无评论

Improve the quality of XML clustering by boosting theory

引用

Journal of computational Information Systems 2013年第11期9卷 4247-4254页

作者： Liu, Zhaojun Li, Xiongfei Li, Wei Key Laboratory of Symbolic Computation and Knowledge Engineering for Ministry of Education Jilin University Changchun 130012 China

XML documents cluster analysis is a hot research topic. Researchers proposed a number of methods to cluster XML document collections. Boosting is successful well-known methods for improving the quality of clustering. This paper proposed a method named ICBQ which improve the clustering quality by boosting theory. The experiment runs on the both real dataset and synthetic dataset, we choose the Nierman's method, Dalamagas's method and Flesca's method to evaluate the ability of ICBQ. The experimental result shows that our approach is more precise than no boosting methods. © 2013 by Binary Information Press.

关键词： XML

来源：评论

学校读者我要写书评

暂无评论

Applying functional dependencies to improve the performance and robustness of hidden naive bayes

引用

Journal of computational Information Systems 2013年第19期9卷 7967-7974页

作者： Qin, Shidong Wang, Limin Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun 130012 China

Hidden Naive Bayes (HNB) has demonstrated remarkable progress in classification accuracy, accurate class probability estimation and ranking. Since HNB is based on one-dependence estimators to get the approximate value of conditional probability, in this paper we try to get more accurate approximations of Bayesian classifier by applying the data dependence theory of relational database. Originally HNB uses a hidden parent of an attribute to represent the influences of all other attributes, while the influence is directed in the Bayesian network. In this paper, we proposed to use the chain rule to estimate the influence among attributes. Compared to HNB, experimental results show that our algorithm has both advantages in classification accuracy and ranking performance which is measured by the area under the ROC curve (AUC). © 2013 Binary Information Press.

关键词： Bayesian networks

来源：评论

学校读者我要写书评

暂无评论

Learning Hierarchical User Interest Models from Web Pages

引用

Wuhan University Journal of Natural Sciences 2006年第1期11卷 6-10页

作者： YANG Feng-qin SUN Tie-li SUN Ji-gui College of Computer Science and Technology Jilin University Changchun 130012 Jilin China Key Laboratory for Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun 120012 Jilin China Department of Computer Science Northeast Normal University Changchun 130024 Jilin China

We propose an algorithm for learning hierarchical user interest models according to the Web pages users have browsed. In this algorithm, the interests of a user are represented into a tree which is called a user interest tree, the content and the structure of which can change simultaneously to adapt to the changes in a user＇s interests. This expression represents a user＇s specific and general interests as a continuurn. In some sense, specific interests correspond to shortterm interests, while general interests correspond to longterm interests. So this representation more really reflects the users＇ interests. The algorithm can automatically model a us er＇s multiple interest domains, dynamically generate the in terest models and prune a user interest tree when the number of the nodes in it exceeds given value. Finally, we show the experiment results in a Chinese Web Site.

关键词： personalization user interest model vector space model agglomerate clustering method

来源：评论

学校读者我要写书评

暂无评论

Attribution rollout: a new way to interpret visual transformer

引用

Journal of Ambient Intelligence and Humanized Computing 2023年第1期14卷 163-173页

作者： Xu, Li Yan, Xin Ding, Weiyue Liu, Zechao College of Computer Science and Technology Harbin Engineering University Nantong Street Heilongjiang Harbin150001 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Qianjin Street Jilin Changchun130012 China Department of Medicine Harvard Medical School Longwood Avenue BostonMA02115 United States

Transformer-based models are dominating the field of natural language processing and are becoming increasingly popular in the field of computer vision. However, the black box characteristics of transformers seriously hamper their application in certain fields. Prior work relies on the raw attention scores or employs heuristic propagation along with the attention graph. In this work, we propose a new way to visualize model. The method computes attention scores based on attribution and then propagates these attention scores through the layers. This propagation involves attention layers and multi-head attention mechanism. Our method extracts salient dependencies in each layer to visualize prediction results. We benchmark our method on recent visual transformer networks and demonstrate its many advantages over the existing interpretability methods. Our code is available at: https://***/yxheartipp/attr-rollout. © 2022, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

A frequency and two-hop configuration checking-driven local search algorithm for the minimum weakly connected dominating set problem

引用

Neural Computing and Applications 2024年第22期36卷 13833-13852页

作者： Li, Ruizhi He, Jintao Lin, Cuisong Liu, Ying Hu, Shuli Yin, Minghao Jilin University of Finance and Economics Changchun130117 China Northeast Normal University Changchun130117 China Business Big Data Research Center of Jilin Province Changchun130117 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China

The minimum weakly connected dominating set problem is a typical NP-hard problem with a wide range of applications. To solve this problem, we propose a frequency property and two-hop configuration checking strategy-driven local search algorithm (FCC2LS). In this algorithm, we first propose a lock-vertex-based initial solution construction procedure. This procedure guarantees that certain vertices, which must be included in the optimal solution, are added to the solution. Second, we propose a two-hop configuration checking strategy and a frequency property. The two-hop configuration checking strategy is a new variant of the original configuration checking strategy, which prohibits vertices with unchanged configuration from being added to the candidate solution. The frequency property is used to record the number of times for each vertex is added to the solution, which can increase the diversity of selected vertices. Third, we combine two scoring functions, Dscore and Nscore, with the above strategies and propose effective vertex selection methods to help the algorithm select suitable vertices to add into or remove from the candidate solutions. Finally, we evaluate the proposed algorithm FCC2LS with four state-of-the-art algorithms on four groups of benchmark instances. Experimental results show that our algorithm performs better on four classical benchmark instances. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Combinatorial optimization

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：