时态图上图模式匹配研究综述  

Graph pattern matching on temporal graphs:A survey

作　　者：李发明[1] 邹兆年[1] 李建中[1] LI Faming;ZOU Zhaonian;LI Jianzhong(Faculty of Computing,Harbin Institute of Technology,Harbin 150001,China)

机构地区：[1]哈尔滨工业大学计算学部,哈尔滨150001

出　　处：《智能计算机与应用》2021年第8期189-192,F0003,共5页Intelligent Computer and Applications

基　　金：国家自然科学基金重点子课题(61532015);国家自然科学基金面上项目(61672189)。

摘　　要：图数据模型是一种通用且已经得到广泛应用的可以处理数据对象间复杂关系的数据模型。然而,现有的大多数研究只关注静态图数据的结构或者顶点和边上的属性,忽略了现实中数据的一个重要特征即时态信息。忽略了时态信息将导致错过很多的有价值的信息,甚至得到错误的信息。作为图研究领域中重要的研究内容之一,图模式匹配问题的研究也需要考虑时态信息。考虑到图模式匹配研究的重要性以及时态信息对数据的重要性,本文根据时态图的快照模型、边流模型和区间模型以及时态图数据的时序性、持续性和演化性对时态图上图模式匹配问题进行了全面地综述,并总结了现有工作的不足。Graph data model is a universal and widely used data model,which can process the complex relationship between data objects.However,most of the existing works on static graphs only focus on the structure or properties of vertices and edges and ignore one important characteristic of data in real life⁃temporal information.It leads to a lot of valuable knowledge being missed or gets wrong results without considering the temporal information.As an important topic in the graph research,the graph pattern matching problem should take temporal information into account as well.Considering the importance of graph pattern matching and the importance of temporal information to the data,this paper thoroughly surveys graph pattern matching on temporal graphs based on the snapshot model,edge-stream model and interval model of the temporal graph,and the temporality,the evolvability and the durability of the temporal data.Finally,the shortcomings of existing works are summarized.

关 键 词：时态信息 时态图 图模式匹配 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]