[1]逯泽锟,于千城,王晓峰,等.基于双重注意力机制的符号网络节点嵌入[J].郑州大学学报(工学版),2023,44(02):68-74.[doi:10.13705/j.issn.1671-6833.2023.02.012]
　HUANG Guoru,YANG Ge,ZENG Bowei,et al.Urban Flood Disaster Control Based on Green-gray-blue Infrastructure Integration[J].Journal of Zhengzhou University (Engineering Science),2023,44(02):68-74.[doi:10.13705/j.issn.1671-6833.2023.02.012]
点击复制

基于双重注意力机制的符号网络节点嵌入()
分享到：

《郑州大学学报(工学版)》[ISSN:1671-6833/CN:41-1339/T]

卷:
44
期数:
2023年02期
页码:
68-74
栏目:
出版日期:
2023-02-27

文章信息/Info

Title:
Urban Flood Disaster Control Based on Green-gray-blue Infrastructure Integration
作者:
逯泽锟1; 2; 于千城1; 2; 王晓峰1; 李 霞1; 2; 王金云3
1. 北方民族大学 计算机科学与工程学院,宁夏 银川 750021; 2. 北方民族大学 图形图像国家民委重点实验室, 宁 夏 银川 750021;3. 北方民族大学 商学院,宁夏 银川 750021

Author(s):
HUANG Guoru1; 2; YANG Ge1; 2; ZENG Bowei1;  LYU Yongpeng1; 2;  REN Xinxin3
1.School of Computer Science and Engineering, Northern Nationalities University, Yinchuan 750021 in Ningxia, Key Laboratory of the National Committee of the National Committee of the University of Nationalities, Ningxia Yinchuan 750021, 2.School of Computer Science and Engineering, Northern National University, Ningxia Yinchuan 750021, 3.Northern National University Business School, Ningxia Yinchuan, Ningxia Yinchuan, Ningxia 750021

关键词:
符号网络;  图神经网络;  图注意力网络;  网络嵌入;  链路预测
Keywords:
signed network;  graph neural network;  graph attention networks;  network embedding;  link prediction
分类号:
TP181
DOI:
10.13705/j.issn.1671-6833.2023.02.012
文献标志码:
A
摘要:
网络节点嵌入是将网络中的节点映射为低维的向量表示,从而可以直接应用基于向量空间的学习方法来 处理链路预测等下游任务。 现有的网络节点嵌入模型大多针对无符号网络,不能直接用于处理符号网络( 通常需 要将符号网络转换成无符号网络进行处理,因而丢弃了边上的正负号所蕴含着的大量有价值的信息) 。 基于图神 经网络(GNNs)提出了一种可以直接处理符号网络的节点嵌入模型,即基于双重注意力机制的符号网络节点嵌入 ( SNEDA) 。 依据结构平衡理论和地位理论,将节点间的路径按照方向和边上的正负信息划分成 20 种不同的模体 (motif)结构。 设计了包含 2 层注意力机制的网络传播模型,当汇聚节点 i 的直接邻居信息时,通过节点级注意力 机制捕获不同邻居节点对节点 i 的向量表示的贡献和影响;当汇聚节点 i 的二阶及二阶以上各阶邻居信息时,用路 径级注意力捕获不同 motif 对节点 i 的向量表示。 通过引入两层注意力机制综合考虑节点层面和路径层面的不同 贡献和影响,不仅提高了算法的时间效率,而且使得最终得到节点 i 的向量表示更有利于提高下游链路预测任务的 预测准确性。 在 4 个真实的社交网络数据集上进行实验,与基准模型相比,SNEDA 模型在 AUC 和 F1 指标上分别 提高了约 3. 1%和 1. 1%。 结果表明该模型得到的网络表示有助于提高链路预测的准确性。
Abstract:
Network node embedding is mapping nodes in a network to a low-dimensional vector representation, so that vector space-based learning methods can be directly applied to handle downstream tasks such as link prediction. Most of the existing network node embedding models were for unsigned networks and could not be directly used to deal with signed networks (usually need to be converted into unsigned networks for processing, thus discarding a lot of valuable information embedded in the positive and negative signs on the edges).A node embedding model (SNEDA)based on graphical neural networks was proposed that could directly deal with symbolic networks. Based on structural balance theory and status theory, the paths between nodes were divided into 20 different motif structures according to the direction and the positive and negative information on the edges. A network propagation model was designed with two levels of attention mechanism, which could capture the contribution and influence of different neighboring nodes to the vector representation of node i by node-level attention mechanism when aggregating the direct neighboring information of node i, and captured the vector representation of different motif to node i by path-level attention when aggregating the second-order and higher-order neighboring information of node i. A two-level attention mechanism was introduced to integrate different contributions and influences at the node level and path level, which could it not only improve the time efficiency of the algorithm but also make the final vector representation of node i more beneficial to improve the prediction accuracy of the downstream link prediction task. Through experiments conducted on four real social network datasets, the SNEDA model improved the AUC and F1 metrics by about 3.1% and 1.1%, respectively, compared with the benchmark model, and the results showed that the network representation obtained by the model could improve the accuracy of link prediction.

参考文献/References:

[1] 任磊, 杜一, 马帅, 等. 大数据可视分析综述[ J] . 软 件学报, 2014, 25(9) : 1909-1936.

 REN L, DU Y, MA S, et al. Visual analytics towards big data[J]. Journal of Software, 2014, 25(9): 1909-1936. 
[2] BORSBOOM D, DESERNO M K, RHEMTULLA M, et al. Network analysis of multivariate data in psychological science[ EB / OL] . ( 2021- 08- 19) [ 2022- 03- 16] . https: / / doi. org / 10. 1038 / s43586-021-00055-w. 
[3] ZHOU J Y, LIU L, WEI W Q, et al. Network representation learning: from preprocessing, feature extraction to node embedding[ J] . ACM Computing Surveys, 2023, 55 (2) : 1-35. 
[4] CUI P, WANG X, PEI J, et al. A survey on network embedding [ J ] . IEEE Transactions on Knowledge and Data Engineering, 2019, 31(5) : 833-852. 
[5] 涂存超, 杨成, 刘知远, 等. 网络表示学习综述[ J] . 中国科学: 信息科学, 2017, 47(8) : 980-996. 
TU C C, YANG C, LIU Z Y, et al. Network representation learning: an overview [ J] . Scientia Sinica Informationis, 2017, 47(8) : 980-996. 
[6] 高岳林, 杨钦文, 王晓峰, 等. 新型群体智能优化算 法综述[ J] . 郑州大学学报( 工学版) , 2022, 43( 3) : 21-30. 
GAO Y L, YANG Q W, WANG X F, et al. Overview of new swarm intelligent optimization algorithms[ J] . Journal of Zhengzhou University ( Engineering Science) , 2022, 43(3) : 21-30.
 [7] 郑建兴, 郭彤彤, 申利华, 等. 基于评论文本情感注 意力的推荐方法研究[ J] . 郑州大学学报( 工学版) , 2022, 43(2) : 44-50, 57. 
ZHENG J X, GUO T T, SHEN L H, et al. Research on recommendation method based on sentimental attention of review text[ J] . Journal of Zhengzhou University ( Engineering Science) , 2022, 43(2) : 44-50, 57. 
[8] 程苏琦, 沈 华 伟, 张 国 清, 等. 符 号 网 络 研 究 综 述 [ J] . 软件学报, 2014, 25(1) : 1-15. CHENG S Q, SHEN H W, ZHANG G Q, et al. Survey of signed network research [ J ] . Journal of Software, 2014, 25(1) : 1-15. 
[9] CHEN J, ZHONG M, LI J, et al. Effective deep attributed network representation learning with topology adapted smoothing[ J] . IEEE Transactions on Cybernetics, 2022, 52(7) : 5935-5946.
 [10] WU Z H, PAN S R, CHEN F W, et al. A comprehensive survey on graph neural networks[ J] . IEEE Transactions on Neural Networks and Learning Systems, 2021, 32(1) : 4-24.
 [11] ZHANG Z W, CUI P, ZHU W W. Deep learning on graphs: a survey [ J] . IEEE Transactions on Knowledge and Data Engineering, 2022, 34(1) : 249-270. 
[12] HEIDER F. Attitudes and cognitive organization[ J] . The Journal of Psychology, 1946, 21(1) : 107-112. 
[13] 刘苗苗, 扈庆翠, 郭景峰, 等. 符号网络链接预测算 法研究综述[ J] . 计算机科学, 2020, 47(2) : 21-30. 
LIU M M, HU Q C, GUO J F, et al. Survey of link prediction algorithms in signed networks[ J] . Computer Science, 2020, 47(2) : 21-30. 
[14] ANCHURI P, MAGDON-ISMAIL M. Communities and balance in signed networks: a spectral approach [ C] / / 2012 IEEE / ACM International Conference on Advances in Social Networks Analysis and Mining. Piscataway: IEEE,2012: 235-242. 
[15] LESKOVEC J, HUTTENLOCHER D, KLEINBERG J. Predicting positive and negative links in online social networks [C] / / Proceedings of the 19th International Conference on World Wide Web. New York: ACM, 2010: 641-650.
 [16] PEROZZI B, AL-RFOU R, SKIENA S. DeepWalk: online learning of social representations [ C] / / Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2014: 701-710. 
[17] WANG S H, TANG J L, AGGARWAL C, et al. Signed network embedding in social media [ C] / / Proceedings of the 2017 SIAM International Conference on Data Mining. Philadelphia: Society for Industrial and Applied Mathematics, 2017: 327-335. 
[18] DERR T, MA Y, TANG J L. Signed graph convolutional networks [ C ] / / 2018 IEEE International Conference on Data Mining. Piscataway: IEEE,2018: 929-934. 
[19] LI Y, TIAN Y, ZHANG J W, et al. Learning signed network embedding via graph attention [ J] . Proceedings of the AAAI Conference on Artificial Intelligence, 2020, 34 (4) : 4772-4779. 
[20] HUANG J J, SHEN H W, HOU L, et al. Signed graph attention networks [ C ] / / Artificial Neural Networks and Machine Learning-ICANN 2019: Workshop and Special Sessions. Cham: Springer, 2019: 566-577. 
[21] HUANG J J, SHEN H W, HOU L, et al. SDGNN: learning node representation for signed directed networks [ J] . Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(1) : 196-203.

相似文献/References:

[1]李文举,姬倩倩,沙利业,等.基于图游走和图注意力的点云分类与分割[J].郑州大学学报(工学版),2024,45(02):33.[doi:10.13705/j.issn.1671-6833.2024.02.006]
　LI Wenju,JI Qianqian,SHA Liye,et al.Point Cloud Classification and Segmentation Based on Graph Walk and Graph Attention[J].Journal of Zhengzhou University (Engineering Science),2024,45(02):33.[doi:10.13705/j.issn.1671-6833.2024.02.006]
[2]刘慧林,范瑞明,程大闯,等.基于图神经网络的智能电网运行状态分析与评估[J].郑州大学学报(工学版),2024,45(06):122.[doi:10.13705/j.issn.1671-6833.2024.06.017]
　LIU Huilin,FAN Ruiming,CHENG Dachuang,et al.Analysis and Evaluation Model of Smart Grid Operation State Basedon Graph Neural Network[J].Journal of Zhengzhou University (Engineering Science),2024,45(02):122.[doi:10.13705/j.issn.1671-6833.2024.06.017]

更新日期/Last Update: 2023-02-25