Heterogeneous Information Networks: the Past, the Present, and the Future
Heterogeneous Information Networks: the Past, the Present, and the Future
复制标题
DOI：
10.14778/3554821.3554901
复制
发表时间：
2022-08
期刊：
影响因子：
--
通讯作者：
Yizhou Sun;Jiawei Han;Xifeng Yan;Philip S. Yu;Tianyi Wu
中科院分区：
文献类型：
--
作者：
Yizhou Sun;Jiawei Han;Xifeng Yan;Philip S. Yu;Tianyi Wu
2011年，我们提出了PathSim来系统地定义和计算异构信息网络（HIN）中节点之间的相似性，其中节点和链接来自不同类型。在PathSim论文中，我们首次引入了具有一般网络模式的HIN，并提出了元路径的概念来系统地定义节点之间的新关系类型。在本文中，我们总结了PathSim论文在学术界和工业界的影响。我们从基于元路径的特征工程的算法开始，然后转向异构网络表示学习的最新发展，包括浅网络嵌入和异构图神经网络。最后，我们将知识图和HIN联系起来，讨论了元路径在符号推理场景中的含义。最后，我们指出了几个未来的发展方向。
In 2011, we proposed PathSim to systematically define and compute similarity between nodes in a heterogeneous information network (HIN), where nodes and links are from different types. In the PathSim paper, we for the first time introduced HIN with general network schema and proposed the concept of meta-paths to systematically define new relation types between nodes. In this paper, we summarize the impact of PathSim paper in both academia and industry. We start from the algorithms that are based on meta-path-based feature engineering, then move on to the recent development in heterogeneous network representation learning, including both shallow network embedding and heterogeneous graph neural networks. In the end, we make the connection between knowledge graphs and HINs and discuss the implication of meta-paths in the symbolic reasoning scenario. Finally, we point out several future directions.