Information cascade prediction of complex networks based on physics-informed graph convolutional network

被引:9
|
作者
Yu, Dingguo [1 ,2 ]
Zhou, Yijie [1 ,2 ]
Zhang, Suiyu [1 ,2 ]
Li, Wenbing [3 ]
Small, Michael [4 ]
Shang, Ke-ke [5 ]
机构
[1] Commun Univ Zhejiang, Intelligent Media Inst, Hangzhou 310018, Peoples R China
[2] Key Lab Film & TV Media Technol Zhejiang Prov, Hangzhou 310018, Peoples R China
[3] Commun Univ Zhejiang, Zhejiang Res Ctr Commun & Cultural Ind, Hangzhou 310018, Peoples R China
[4] Univ Western Australia, Dept Math & Stat, Complex Syst Grp, Perth 6009, Australia
[5] Nanjing Univ, Computat Commun Collaboratory, Nanjing, Peoples R China
来源
NEW JOURNAL OF PHYSICS | 2024年 / 26卷 / 01期
基金
中国国家自然科学基金; 中国国家社会科学基金;
关键词
information cascade prediction; physics-informed network; graph convolutional network; complex networks;
D O I
10.1088/1367-2630/ad1b29
中图分类号
O4 [物理学];
学科分类号
0702 ;
摘要
Cascade prediction aims to estimate the popularity of information diffusion in complex networks, which is beneficial to many applications from identifying viral marketing to fake news propagation in social media, estimating the scientific impact (citations) of a new publication, and so on. How to effectively predict cascade growth size has become a significant problem. Most previous methods based on deep learning have achieved remarkable results, while concentrating on mining structural and temporal features from diffusion networks and propagation paths. Whereas, the ignorance of spread dynamic information restricts the improvement of prediction performance. In this paper, we propose a novel framework called Physics-informed graph convolutional network (PiGCN) for cascade prediction, which combines explicit features (structural and temporal features) and propagation dynamic status in learning diffusion ability of cascades. Specifically, PiGCN is an end-to-end predictor, firstly splitting a given cascade into sub-cascade graph sequence and learning local structures of each sub-cascade via graph convolutional network , then adopting multi-layer perceptron to predict the cascade growth size. Moreover, our dynamic neural network, combining PDE-like equations and a deep learning method, is designed to extract potential dynamics of cascade diffusion, which captures dynamic evolution rate both on structural and temporal changes. To evaluate the performance of our proposed PiGCN model, we have conducted extensive experiment on two well-known large-scale datasets from Sina Weibo and ArXIv subject listing HEP-PH to verify the effectiveness of our model. The results of our proposed model outperform the mainstream model, and show that dynamic features have great significance for cascade size prediction.
引用
收藏
页数:14
相关论文
共 50 条
  • [1] Probabilistic Physics-Informed Graph Convolutional Network for Active Distribution System Voltage Prediction
    Su, Tong
    Zhao, Junbo
    Pei, Yansong
    Ding, Fei
    IEEE TRANSACTIONS ON POWER SYSTEMS, 2023, 38 (06) : 5969 - 5972
  • [2] Prediction of subsonic cascade flow fields by physics-informed graph neural networks with unbalanced data
    Feng, Yunyang
    Yuan, Wei
    Song, Xizhen
    Yan, Zhaoqi
    Pan, Tianyu
    Aerospace Science and Technology, 2025, 162
  • [3] HCP-PIGN: Efficient heat conduction prediction by physics-informed graph convolutional neural network
    Peng, Jiang-Zhou
    Aubry, Nadine
    Li, Yu-Bai
    Chen, Zhi-Hua
    Mei, Mei
    Hua, Yue
    INTERNATIONAL JOURNAL OF HEAT AND FLUID FLOW, 2024, 109
  • [4] Physics-Informed Graph Convolutional Recurrent Network for Cyber-Attack Detection in Chemical Process Networks
    Wu, Guoquan
    Zhang, Haohao
    Wu, Wanlu
    Wang, Yujia
    Wu, Zhe
    INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH, 2025, 64 (06) : 3370 - 3382
  • [5] Physics-informed neural networks based cascade loss model
    Feng Y.
    Song X.
    Yuan W.
    Lu H.
    Hangkong Dongli Xuebao/Journal of Aerospace Power, 2023, 38 (07): : 845 - 855
  • [6] A High-Efficient Hybrid Physics-Informed Neural Networks Based on Convolutional Neural Network
    Fang, Zhiwei
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2022, 33 (10) : 5514 - 5526
  • [7] Physics-informed graph convolutional neural network for modeling fluid flow and heat convection
    Peng, Jiang-Zhou
    Hua, Yue
    Li, Yu-Bai
    Chen, Zhi-Hua
    Wu, Wei-Tao
    Aubry, Nadine
    PHYSICS OF FLUIDS, 2023, 35 (08)
  • [8] Data and physics-driven modeling for fluid flow with a physics-informed graph convolutional neural network
    Peng, Jiang -Zhou
    Hua, Yue
    Aubry, Nadine
    Chen, Zhi-Hua
    Mei, Mei
    Wu, Wei-Tao
    OCEAN ENGINEERING, 2024, 301
  • [9] Fracture network flow prediction with uncertainty using physics-informed graph features
    Strait, Justin D.
    Moran, Kelly R.
    Hyman, Jeffrey D.
    Viswanathan, Hari S.
    Sweeney, Matthew R.
    Stauffer, Philip H.
    COMPUTATIONAL GEOSCIENCES, 2023, 27 (06) : 1111 - 1132
  • [10] Fracture network flow prediction with uncertainty using physics-informed graph features
    Justin D. Strait
    Kelly R. Moran
    Jeffrey D. Hyman
    Hari S. Viswanathan
    Matthew R. Sweeney
    Philip H. Stauffer
    Computational Geosciences, 2023, 27 : 1111 - 1132