摘要
作为深度神经网络向非欧式数据上的扩展,图神经网络(GNN)已经在图节点分类任务、链接预测任务和图分类任务中取得了显著成就。在图分类任务上,当前方法一般通过层次化的池化过程同时考虑图的局部和全局结构信息以学习高层次的图表示。在对当前的图分类模型进行对比分析后,考虑当前方法的不足,结合不同方法的优势,提出结构和特征融合池化模型(SAFPool)。SAFPool模型在池化时使用了两个聚类分配矩阵生成模块,分别是基于结构的聚类学习和基于特征的聚类学习模块,基于结构的聚类学习根据图结构信息对结构相似的节点聚类,基于特征的聚类学习则根据图节点特征对特征相似的节点聚类。二者的聚类结果加权聚合后便能获取实现聚类策略的聚类分配矩阵以同时利用图结构和节点特征信息。最后,在多个图分类数据集上通过对比实验和可视化说明了同时显式地利用图节点特征信息和图结构信息实现聚类策略的有效性。
Graph neural networks(GNN), which extend deep neural networks to non-Euclidean data, have been proven to be powerful for numerous graph related tasks such as node classification, link prediction, and graph classification.On the task of graph classification, recent studies aim to learn graph-level representation through a hierarchical pooling procedure using the local and global structure information of the graph. After comparing and analyzing the current graph classification model, this paper proposes the structure and feature fusion pooling model(SAFPool) considering shortcomings of the current method and combining the advantages of different methods. SAFPool utilizes two assignment matrix generation modules during the pooling process, which are structure-based cluster learning and feature-based cluster learning modules. Structure-based cluster learning module clusters nodes with similar structures based on graph structure information, and feature-based cluster learning clusters nodes with similar features based on graph node features. Then the two clustering assignment matrices are weighted and aggregated to obtain the clustering assignment matric which implements the clustering strategy to utilize graph structure and node feature information at the same time. Finally, comparative experiments and visualization on multiple graph classification datasets demonstrate the effectiveness of using graph node information and structure information to implement a clustering strategy in graph classification.
作者
马涪元
王英
李丽娜
汪洪吉
MA Fuyuan;WANG Ying;LI Lina;WANG Hongji(College of Computer Science and Technology,Jilin University,Changchun 130012,China;Key Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education(Jilin University),Changchun 130012,China;College of Artificial Intelligence,Jilin University,Changchun 130012,China)
出处
《计算机科学与探索》
CSCD
北大核心
2023年第1期179-186,共8页
Journal of Frontiers of Computer Science and Technology
基金
国家自然科学基金(61872161)
中国博士后科学基金(2017M611301)
吉林省自然科学基金(20200201297JC,2018101328JC)
吉林省发展和改革基金(2019C053-8)
吉林省教育委员会基金(JJKH20191257KJ)
吉林大学学科交叉融合创新项目(419021421615,JLUXKJC2020207)。
关键词
图神经网络
图分类
图池化
聚类分配矩阵
层次化模型
graph neural network
graph classification
graph pooling
clustering assignment matrix
hierarchical model
