注意力感知的边-节点交换图神经网络模型

doi:10.11959/j.issn.1000-0801.2024017

Abstract

Abstract:

An attention aware edge-node exchange graph neural network (AENN) model was proposed, which used edge-node switched convolutional graph neural network method for graph encoding in a graph structured data representation framework for semi supervised classification and regression analysis.AENN is an universal graph encoding framework for embedding graph nodes and edges into a unified latent feature space.Specifically, based on the original undirected graph, the convolution between edges and nodes was continuously switched, and during the convolution process, attention mechanisms were used to assign weights to different neighbors, thereby achieving feature propagation.Experimental studies on three datasets show that the proposed method has significant performance improvements in semi-supervised classification and regression analysis compared to existing methods.

Key words: graph neural network, message passing, attention mechanism, hypergraph, line graph

CLC Number:

TP393

Ruiqin WANG, Yimin HUANG, Qishun JI, Chaoyi WAN, Zhifeng ZHOU. Attention aware edge-node exchange graph neural network[J]. Telecommunications Science, 2024, 40(1): 106-114.

Figures/Tables 9

References 12

[1]	CHEN C , ZHANG M , ZHANG Y F ,et al. Efficient neural matrix factorization without sampling for recommendation[J]. ACM Transactions on Information Systems, 2020,38(2): 1-28.
[2]	CHEN L , WU L , HONG R C ,et al. Revisiting graph based collaborative filtering:a linear residual graph convolutional network approach[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2020,34(1): 27-34.
[3]	CHEN M , WEI Z W , HUANG Z F ,et al. Simple and deep graph convolutional networks[J]. arXiv preprint, 2020,arXiv:2007.02133.
[4]	XIA X , YIN H , YU J ,et al. Self-supervised hypergraph convolutional networks for session-based recommendation[C]// Proceedings of the AAAI conference on artificial intelligence. 2021,35(5): 4503-4511.
[5]	WU J C , WANG X , FENG F L ,et al. Self-supervised graph learning for recommendation[C]// Proceedings of the Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York:ACM Press, 2021: 726-735.
[6]	YU W H , QIN Z . Graph convolutional network for recommendation with low-pass collaborative filters[C]// Proceedings of the Proceedings of the 37th International Conference on Machine Learning. New York:ACM Press, 2020: 10936-10945.
[7]	YU W H , QIN Z . Sampler design for implicit feedback data by noisy-label robust learning[C]// Proceedings of the Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York:ACM Press, 2020: 861-870.
[8]	SUN J N , GUO W , ZHANG D C ,et al. A framework for recommending accurate and diverse items using Bayesian graph convolutional neural networks[C]// Proceedings of the Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery ＆ Data Mining. New York:ACM Press, 2020: 2030-2039.
[9]	JI S Y , FENG Y F , JI R R ,et al. Dual channel hypergraph collaborative filtering[C]// Proceedings of the Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery ＆ Data Mining. New York:ACM Press, 2020: 2020-2029.
[10]	WANG M H , LIN Y J , LIN G L ,et al. M2GRL:a multi-task multi-view graph representation learning framework for web-scale recommender systems[C]// Proceedings of the Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery ＆ Data Mining. New York:ACM Press, 2020: 2349-2358.
[11]	ZHANG H , LIU Q , LIU Z . Simplification of graph convolutional networks:a matrix factorization-based perspective[J]. arXiv preprint, 2020,arXiv:2007.09036.
[12]	MAO K , ZHU J , XIAO X ,et al. UltraGCN:ultra simplification of graph convolutional networks for recommendation[J]. arXiv preprint, 2021,arXiv:2110.15114.

Metrics

Recommended 0

No Suggested Reading articles found!

数据集	标签比例	GCN	GAT	GraphSAGE	AENN
Cora	3%	74%±2.8%	74.8%±7.9%	64.2%±4.0%	76.4%±1.1%
	1%	61.0%±7.2%	68.6%±8.0%	49.0%±5.8%	64.1%±0.9%
	0.5%	52.9%±7.4%	54.4%±6.9%	37.5%±5.4%	54.7%±2.9%
Citeseer	1%	58.3%±4.0%	59.5%±4.3%	51.0%±5.7%	63.5%±1.5%
	0.5%	47.7%±4.4%	48.2%±4.1%	33.8%±7.0%	50.6%±3.0%
	0.3%	39.2%±6.3%	40.9%±6.9%	25.7%±6.1%	45.4%±3.6%
PubMed	0.1%	63.0%±5.5%	65.6%±5.5%	65.4%±6.2%	70.0%±2.1%
	0.05%	64.6%±7.5%	65.4%±9.7%	53.0%±8.0%	66.7%±1.2%
	0.03%	57.9%±8.1%	60.9%±8.8%	45.4%±5.5%	62.4%±2.4%

数据集	基准模型	+Attention	+边图	AENN
Cora	65.7%±2.1%	70.4%±2.8%	71.5%±2.1%	73.4%±1.1%
Citeseer	57.4%±3.3%	60.3%±2.1%	59.7%±2.4%	61.5%±1.5%
PubMed	66.2%±4.6%	68.8%±3.7%	68.0%±3.4%	70.0%±2.1%

[1]	Yong ZHANG, Haoshuang LIU, Qi ZHANG, Wenzhe LIU. Multi-adversarial domain adaptation method based on feature correction [J]. Telecommunications Science, 2024, 40(1): 71-82.
[2]	Yong ZHANG, Jikui LIU, Wenlong KE. EEG emotion recognition based on parallel separable convolution and label smoothing regularization [J]. Telecommunications Science, 2023, 39(5): 116-128.
[3]	Jianfeng DAI, Xingyu CHEN, Ligang DONG, Xian JIANG. A triple joint extraction method combining hybrid embedding and relational label embedding [J]. Telecommunications Science, 2023, 39(2): 132-144.
[4]	Bei YANG, Xin LIANG, Hang YIN, Zheng JIANG, Xiaoming SHE. Self-attention mechanism-based CSI eigenvector feedback for massive MIMO [J]. Telecommunications Science, 2023, 39(11): 128-136.
[5]	Shang WU, Lei SHEN, Lijun WANG, Ruxu ZHANG, Xin HU. Ultrashort wave satellite channel classification and recognition algorithm based on mirror filled spectrum and LA-ResNet50 [J]. Telecommunications Science, 2023, 39(10): 74-84.
[6]	Le YU, Lidong ZHU, Liang JIN, Jiali LI, Yilun LIU, Cheng GUO. Generalized approximate message passing based SrCMA satellite receiver [J]. Telecommunications Science, 2022, 38(4): 77-89.
[7]	Yijiang LI, Huibiao YE, Renhua XIE, Jiali LOU, Danna ZHUANG, Chuanhuang LI. Intelligent prediction method of network performance based on graph neural network [J]. Telecommunications Science, 2022, 38(3): 143-157.
[8]	Yong SONG, Zhiwei YAN, Yukun QIN, Dongming ZHAO, Xiaozhou YE, Yuanyuan CHAI, Ye OUYANG. Customer service complaint work order classification based on matrix factorization and attention multi-task learning [J]. Telecommunications Science, 2022, 38(2): 103-110.
[9]	Nan JIN, Ruiqin WANG, Yuecong LU. Ebbinghaus forgetting curve and attention mechanism based recommendation algorithm [J]. Telecommunications Science, 2022, 38(10): 89-97.
[10]	Jinyang SUN, Baisong LIU, Hao REN, Jiangbo QIAN. TAGAN: an academic paper adversarial recommendation algorithm incorporating fine-grained semantic features [J]. Telecommunications Science, 2021, 37(8): 57-65.
[11]	Xiaoqun SONG, Ming JIN, Zhongjie JIA. Joint user activity and signal detection for massive multiple-input multiple-output [J]. Telecommunications Science, 2021, 37(5): 113-123.
[12]	Jinhua WANG,Na YING,Chendu ZHU,Zhaosen LIU,Zhedong CAI. Speech emotion recognition algorithm based on spectrogram feature extraction of deep space attention feature [J]. Telecommunications Science, 2019, 35(7): 100-108.
[13]	Yiming WANG,Ken CHEN,Aihaiti ABUDUSALAMU. End-to-end audiovisual speech recognition based on attention fusion of SDBN and BLSTM [J]. Telecommunications Science, 2019, 35(12): 79-89.
[14]	Hongyang ZHANG,Changliang ZHENG,Yu LU,Huaping LIU. Impact of overload factor on uplink SCMA system [J]. Telecommunications Science, 2018, 34(6): 64-71.
[15]	Mingming JI,Jianhong ZHENG. Research on low complexity decoding for SCMA based on serial strategies of variable node [J]. Telecommunications Science, 2017, 33(9): 50-57.

Attention aware edge-node exchange graph neural network

RichHTML

PDF下载

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 9

References 12

Related Articles 15

Metrics

Recommended 0