A node importance prediction algorithm based on graph attention and contrastive learning

doi:10.1088/1674-1056/ae0680

Abstract In complex network analysis, node ranking is vital for propagation prediction, structural optimization, and intervention strategy design, yet existing methods often fail to effectively integrate community information in dynamic settings. To address this, this paper proposes a node ranking method that combines graph attention mechanisms with contrastive learning. Community detection is employed to extract node-level community features, and a joint embedding module is designed to fuse global and local structures, thereby incorporating community information into node representations. Based on this, a multi-layer graph attention network adaptively learns node and neighborhood features, while contrastive learning mitigates interference from dynamic evolution and strengthens the model's ability to capture multi-scale structural differences. Experiments on multiple dynamic network datasets show that the proposed method significantly outperforms existing approaches in ranking accuracy, particularly in networks with higher average degrees and clearer community structures. These results validate the effectiveness of the method in enhancing feature representation and modeling multi-scale dynamic node influence.

Keywords: node importance community features graph attention contrastive learning

Received: 12 August 2025
Revised: 09 September 2025
Accepted manuscript online: 15 September 2025

PACS:	89.75.Hc	(Networks and genealogical trees)
	89.20.Ff	(Computer science and technology)
	05.10.-a	(Computational methods in statistical physics and nonlinear dynamics)

Corresponding Authors: Yuming Zhang, Zhan Su
E-mail: zym1013@163.com;suzhan@usst.edu.cn

Cite this article:

Jun Ai(艾均), Yuming Zhang(张玉明), Zhan Su(苏湛), Chenye Guo(郭晨晔), and Mingsong Li(李铭松) A node importance prediction algorithm based on graph attention and contrastive learning 2026 Chin. Phys. B 35 058901

[1] Battiston F, Cencetti G, Iacopini I, et al. 2020 Phys. Rep. 874 1
[2] Fan C, Zeng L, Sun Y, et al. 2020 Nature Machine Intelligence 2 317
[3] Guo S, Zhou D, Fan J, et al. 2019 EPJ Data Science 8 28
[4] Chakrapani H B, Chourasia S, Gupta S, et al. 2021 Computers in Biology and Medicine 133 104378
[5] Fang X and Hu P J H 2018 MIS Quarterly 42 63
[6] C F L 1978 Social Networks 1 215
[7] Chen D, Lü L, Shang M S, et al. 2012 Physica A 391 1777
[8] Namtirtha A, Dutta A and Dutta B 2018 Proc. 10th Int. Conf. Communication Systems and Networks (COMSNETS) pp. 81-88
[9] Freeman L C 1977 Sociometry 40 35
[10] Sabidussi G 1966 Psychometrika 31 581
[11] Bonacich P 1972 Journal of Mathematical Sociology 2 113
[12] Zhao Z, Wang X, Zhang W and Zhu Z 2015 Entropy 17 2228
[13] Ghalmane Z, El Hassouni M, Cherifi C and Cherifi H 2019 EPJ Data Science 8 15
[14] Asgharian Rezaei A, Munoz J, Jalili M and Khayyam H 2023 Expert Systems with Applications 214 119086
[15] Hu G, Xu X, ZhangWM and Zhou Y 2019 Acta Electronica Sinica 47 358
[16] Yu E Y, Wang Y P, Fu Y, et al. 2020 Knowledge-Based Systems 198 105893
[17] Ou Y, Guo Q, Xing J L and Liu J G 2022 Expert Systems with Applications 203 117515
[18] Ahmad W, Wang B and Chen S 2024 Applied Intelligence 54 3260
[19] Blondel V D, Guillaume J L, Lambiotte R and Lefebvre E 2008 J. Stat. Mech.: Theory Exp. 2008 P10008
[20] Veličković P, Cucurull G, Casanova A, Romero A, Liò P and Bengio Y 2018 Proc. Int. Conf. Learning Representations (ICLR)
[21] Ying R, He R, Chen K, Eksombatchai P, Hamilton W L and Leskovec J 2018 Proc. 24th ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining (KDD) pp. 974-983
[22] Cao H, Wei L, Zhou W and Hu S 2024 arXiv 2407.10474
[23] Guo Y, Tang C, Wu H and Chen B 2024 Proc. IEEE Int. Conf. Acoustics, Speech and Signal Processing (ICASSP) pp. 1921-1925
[24] Gutmann M and Hyvärinen A 2010 Proc. 13th Int. Conf. Artificial Intelligence and Statistics (AISTATS) pp. 297-304
[25] van den Oord A, Li Y and Vinyals O 2018 arXiv 1807.03748
[26] Jiang C, et al. 2024 Proc. IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPR) pp. 27036-27045
[27] Sung C, et al. 2024 Proc. IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPR) pp. 3732-3741
[28] Hethcote H W 2000 SIAM Rev. 42 599
[29] Ahmad W, Wang B and Chen S 2024 Appl. Intell. 54 3260
[30] Priebe C E, Conroy J M, Marchette D J and Park Y 2005 Comput. Math. Organ. Theory 11 229
[31] Rahman M and Hasan M A 2016 Mach. Learn. Knowl. Discov. Databases 9851 394
[32] Viswanath B, Mislove A, Cha M and Gummadi K P 2009 Proc. 2nd ACM Workshop on Online Social Networks pp. 37-42
[33] Mikolov T, Chen K, Corrado G and Dean J 2013 arXiv 1301.3781
[34] Barabási A L and Albert R 1999 Science 286 509
[35] Su X, Cheng J, Yang H, Leng M, Zhang W and Chen X 2020 Int. J. Mod. Phys. C 31 2050094
[36] Greene D, Doyle D and Cunningham P 2010 Proc. Int. Conf. Advances in Social Networks Analysis and Mining (ASONAM) pp. 176-183
[37] Wang Z, Zhao Y, Xi J and Du C 2016 Physica A 461 171
[38] Dorogovtsev S N, Goltsev A V and Mendes J F F 2006 Phys. Rev. Lett. 96 040601
[39] Bonacich P 1987 Am. J. Sociol. 92 1170
[40] Xu G and Dong C 2024 Expert Syst. Appl. 235 121154

[1]	Quantum decoder design for subsystem surface code based on multi-head graph attention and edge weighting Nai-Hua Ji(纪乃华), Hui-Qian Sun(孙汇倩), Bo Xiao(肖博), Ping-Li Song(宋平俐), and Hong-Yang Ma(马鸿洋). Chin. Phys. B, 2025, 34(2): 020309.
[2]	AG-GATCN: A novel method for predicting essential proteins Peishi Yang(杨培实), Pengli Lu(卢鹏丽), and Teng Zhang(张腾). Chin. Phys. B, 2023, 32(5): 058902.
[3]	Ranking important nodes in complex networks by simulated annealing Yu Sun(孙昱), Pei-Yang Yao(姚佩阳), Lu-Jun Wan(万路军), Jian Shen(申健), Yun Zhong(钟赟). Chin. Phys. B, 2017, 26(2): 020201.

No Suggested Reading articles found!

Viewed

Full text

Abstract

Cited

Metrics
Related Articles

A node importance prediction algorithm based on graph attention and contrastive learning

Cite this article:

Online attention