---
title: "HyperDet: Source Detection in Hypergraphs via Interactive Relationship Construction and Feature-rich Attention Fusion"
authors:
  - "Le Cheng"
  - "Peican Zhu"
  - "Yangming Guo"
  - "Keke Tang"
  - "Chao Gao"
  - "Zhen Wang"
doi: "10.24963/ijcai.2025/307"
url: "https://ijcai.org/proceedings/2025/307"
pdf: "https://ijcai.org/proceedings/2025/0307.pdf"
venue: "Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence"
year: 2025
pages: "2758-2766"
track: "Main Track"
keywords:
  - "Data Mining: DM: Networks"
  - "Data Mining: DM: Mining graphs"
  - "Machine Learning: ML: Deep learning architectures"
  - "Machine Learning: ML: Convolutional networks"
---

# HyperDet: Source Detection in Hypergraphs via Interactive Relationship Construction and Feature-rich Attention Fusion

Le Cheng, Peican Zhu, Yangming Guo, Keke Tang, Chao Gao, Zhen Wang

*Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence*, pages 2758-2766. DOI: [10.24963/ijcai.2025/307](https://doi.org/10.24963/ijcai.2025/307)

## Abstract

Hypergraphs offer superior modeling capabilities for social networks, particularly in capturing group phenomena that extend beyond pairwise interactions in rumor propagation. Existing approaches in rumor source detection predominantly focus on dyadic interactions, which inadequately address the complexity of more intricate relational structures. In this study, we present a novel approach for Source Detection in Hypergraphs (HyperDet) via Interactive Relationship Construction and Feature-rich Attention Fusion. Specifically, our methodology employs an Interactive Relationship Construction module to accurately model both the static topology and dynamic interactions among users, followed by the Feature-rich Attention Fusion module, which autonomously learns node features and discriminates between nodes using a self-attention mechanism, thereby effectively learning node representations under the framework of accurately modeled higher-order relationships. Extensive experimental validation confirms the efficacy of our HyperDet approach, showcasing its superiority relative to current state-of-the-art methods.