EUVIMEDEUVIMED entdecken
Crossref

Link prediction algorithm based on dual-branch fusion network

Yan Zhang, Zeng Fan Cheng, Li Wang

Physica Scripta · 13.11.2026

DOI: 10.1088/1402-4896/aec206 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Physica Scripta, online erschienen am 13.11.2026. Verfasst von Yan Zhang und 2 weiteren Autor:innen. Gefördert von: Business Modeling and Collaborative Optimization for Industrial Chain Cooperative Enterprises.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Link prediction aims to infer unobserved or potentially existing edges within graph structures. It plays a crucial role in uncovering latent graph information, supplementing missing data, and discovering new knowledge. Graph Attention Networks (GATs) have emerged as a mainstream architecture for modeling complex topological relationships in link prediction, thanks to their ability to dynamically capture the importance of inter-node interactions. However, GATs still suffer from limitations such as ineffective capture of potential relationships among multi-hop neighbors and weak nonlinear fitting capability. To address these issues, this paper proposes a dualbranch link prediction network integrating graph attention and fusion attention. In the improved graph attention branch, the Kolmogorov-Arnold Network (KAN) is introduced, replacing edge weights with learnable nonlinear basis functions and parameterizing the activation function for each edge. This enhancement strengthens the network's ability to fit complex nonlinear relationships and improves its adaptability to high-dimensional graph data. The fusion attention branch combines the global average pooling of SE (Squeeze-and-Excitation) channel attention with the dual pooling operations of CAM (Channel Attention Module) channel attention. This integration leverages SE's global information compression capability and CAM's local salient feature extraction ability, significantly boosting the model's performance in complex scenarios. Experiments on public datasets Yeast, Citeseer, and ADV demonstrate that the proposed algorithm achieves AUC scores of 90.68%, 87.56%, and 88.98% respectively on 80% training data, outperforming baseline methods and effectively enhancing link prediction performance.

Förderung

  • Business Modeling and Collaborative Optimization for Industrial Chain Cooperative Enterprises

Zitieren

APA

Zhang, Y., Cheng, Z. F., & Wang, L. (2026). Link prediction algorithm based on dual-branch fusion network. Physica Scripta. https://doi.org/10.1088/1402-4896/aec206

Vancouver

Zhang Y, Cheng ZF, Wang L. Link prediction algorithm based on dual-branch fusion network. Physica Scripta. 2026. doi:10.1088/1402-4896/aec206

BibTeX
@article{Zhang2026,
  author = {Zhang, Yan and Cheng, Zeng Fan and Wang, Li},
  title = {Link prediction algorithm based on dual-branch fusion network},
  journal = {Physica Scripta},
  year = {2026},
  doi = {10.1088/1402-4896/aec206}
}
Kostenlos starten →