Skip to main content
QUICK REVIEW

[论文解读] Query Driven-Graph Neural Networks for Community Search: From Non-Attributed, Attributed, to Interactive Attributed

Yuli Jiang, Yu Rong|arXiv (Cornell University)|Apr 8, 2021
Advanced Graph Neural Networks被引用 6
一句话总结

本文提出查询驱动图神经网络(QD-GNN)和属性查询驱动图神经网络(AQD-GNN),为新型图神经网络模型,通过整合局部查询感知表征与全局图嵌入以及二部属性图,联合建模依赖于查询的社区结构与节点属性。AQD-GNN在交互式设置下的属性社区搜索任务中实现最先进性能,F1得分相较先前方法提升6.29%。

ABSTRACT

Given one or more query vertices, Community Search (CS) aims to find densely intra-connected and loosely inter-connected structures containing query vertices. Attributed Community Search (ACS), a related problem, is more challenging since it finds communities with both cohesive structures and homogeneous vertex attributes. However, most methods for the CS task rely on inflexible pre-defined structures and studies for ACS treat each attribute independently. Moreover, the most popular ACS strategies decompose ACS into two separate sub-problems, i.e., the CS task and subsequent attribute filtering task. However, in real-world graphs, the community structure and the vertex attributes are closely correlated to each other. This correlation is vital for the ACS problem. In this paper, we propose Graph Neural Network models for both CS and ACS problems, i.e., Query Driven-GNN and Attributed Query Driven-GNN. In QD-GNN, we combine the local query-dependent structure and global graph embedding. In order to extend QD-GNN to handle attributes, we model vertex attributes as a bipartite graph and capture the relation between attributes by constructing GNNs on this bipartite graph. With a Feature Fusion operator, AQD-GNN processes the structure and attribute simultaneously and predicts communities according to each attributed query. Experiments on real-world graphs with ground-truth communities demonstrate that the proposed models outperform existing CS and ACS algorithms in terms of both efficiency and effectiveness. More recently, an interactive setting for CS is proposed that allows users to adjust the predicted communities. We further verify our approaches under the interactive setting and extend to the attributed context. Our method achieves 2.37% and 6.29% improvements in F1-score than the state-of-the-art model without attributes and with attributes respectively.

研究动机与目标

  • 解决现有社区搜索方法依赖刚性、预定义子图模式(如k-core、k-truss)的局限性,这些模式无法捕捉灵活的、依赖于查询的社区结构。
  • 克服属性社区搜索中的属性无关性问题,即属性被独立处理而非建模其隐含相关性。
  • 消除将社区搜索分解为结构与属性过滤两阶段的处理方式,避免结构紧密性与属性同质性之间的解耦。
  • 开发统一的、端到端的图神经网络框架,以响应用户查询,联合优化结构紧密性与属性同质性。
  • 将所提模型扩展至交互式社区搜索,支持用户反馈的迭代优化,提升在真实应用场景中的适应能力。

提出的方法

  • QD-GNN将查询驱动组件与全局图嵌入相结合,利用依赖于查询的注意力机制生成局部社区表征,无需重新训练。
  • AQD-GNN将节点属性建模为节点与属性之间的二部图,使图神经网络能够学习属性之间的关系依赖。
  • 特征融合模块将主图中的结构表征与二部图中的属性表征整合到统一嵌入空间。
  • 模型采用基于阈值的约束BFS将输出logits转换为离散社区集合,γ在验证数据上进行调优。
  • 通过与ICS-GNN集成,将框架扩展至交互式社区搜索,支持基于用户反馈的迭代优化。
  • 模型训练采用预测社区成员关系的交叉熵损失,超参数(dropout率、初始学习率、训练轮数)通过消融实验调优。

实验结果

研究问题

  • RQ1基于图神经网络的模型能否在不依赖预定义子图模式的前提下,联合优化社区搜索中的结构紧密性与属性同质性?
  • RQ2如何有效建模属性之间的隐含相关性,以提升属性社区搜索的质量?
  • RQ3端到端联合学习结构与属性是否优于传统的先结构后属性过滤的两阶段方法?
  • RQ4所提出的图神经网络模型能否有效扩展至交互式社区搜索,实现用户反馈对结果的持续优化?
  • RQ5模型性能与泛化能力对超参数(如阈值γ、训练集大小、dropout率)的敏感性如何?

主要发现

  • 在交互式设置下的属性社区搜索任务中,AQD-GNN相较最先进方法实现6.29%的F1得分提升。
  • 模型对阈值γ具有鲁棒性,在0.3至0.7的广泛范围内表现稳定,表明对超参数调优的敏感度较低。
  • 当训练集大小从50增至100时,性能显著提升,但超过100后趋于稳定,表明在中等数据规模后收益递减。
  • 模型在200个训练轮次内收敛,且在dropout率0.1至0.7范围内保持稳定性能,最优表现出现在0.5处。
  • 消融实验验证,结构与属性的联合学习优于解耦方法,证实了模型设计的有效性。
  • 所提框架在多个真实世界图数据集上,相较于现有方法,在效率与效果方面均表现更优,且具备真实社区标签的验证支持。

更好的研究,从现在开始

从阅读论文到最终审阅,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。