ISSN 1000-1239 CN 11-1777/TP

计算机研究与发展 ›› 2021, Vol. 58 ›› Issue (7): 1436-1455.doi: 10.7544/issn1000-1239.2021.20200685

所属专题: 2021虚假信息检测专题

• 信息处理 • 上一篇    下一篇

基于深度学习的图异常检测技术综述

陈波冯1,李靖东1,卢兴见1,沙朝锋2,王晓玲1,张吉3   

  1. 1(华东师范大学计算机科学与技术学院 上海 200062);2(复旦大学计算机科学技术学院 上海 200433);3(之江实验室 杭州 310000) (51194501030@stu.ecnu.edu.cn)
  • 出版日期: 2021-07-01
  • 基金资助: 
    国家自然科学基金项目(61972155);浙江省自然科学基金重点项目(LZ21F030001);之江实验室PI研究项目(111007-PI2001);之江实验室开放课题资助项目(2019KB0AB04)

Survey of Deep Learning Based Graph Anomaly Detection Methods

Chen Bofeng1, Li Jingdong1, Lu Xingjian1, Sha Chaofeng2, Wang Xiaoling1, Zhang Ji3   

  1. 1(School of Computer Science and Technology, East China Normal University, Shanghai 200062);2(School of Computer Science, Fudan University, Shanghai 200433);3(Zhejiang Lab, Hangzhou 310000)
  • Online: 2021-07-01
  • Supported by: 
    This work was supported by the National Natural Science Foundation of China (61972155), the Zhejiang Provincial Natural Science Foundation of China (LZ21F030001), the PI Research Project of Zhejiang Lab (111007-PI2001), and Zhejiang Lab (2019KB0AB04).

摘要: 图异常检测旨在大图或海量图数据库中寻找“陌生”或“不寻常”模式,具有广泛的应用场景.深度学习可以从数据中学习隐含的规律,在提取数据中潜在复杂模式方面表现出优越的性能.近年来随着基于深度神经网络的图表示学习取得显著进展,如何利用深度学习方法进行图异常检测引起了学术界和产业界的广泛关注.尽管最近一系列研究从图的角度对异常检测技术进行了调研,但是缺少对深度学习技术下的图异常检测技术的关注.首先给出了静态图和动态图上各类常见的异常定义,然后调研了基于深度神经网络的图表示学习方法,接着从静态图和动态图的角度出发,梳理了基于深度学习的图异常检测的研究现状,并总结了图异常检测的应用场景和相关数据集,最后讨论了图异常检测技术目前面临的挑战和未来的研究方向.

关键词: 异常检测, 深度学习, 图网络, 图表示学习, 图神经网络

Abstract: Graph anomaly detection aims to find “strange” or “unusual” patterns in large graph or massive graph databases, and it has a wide range of application scenarios. Deep learning can learn the hidden rules from the data, and it has excellent performance in extracting potential complex patterns from data. With the great development of graph representation learning in recent years, how to detect graph anomaly using deep learning methods has attracted extensive attention in the area of academia and industry. Although a series of recent studies have investigated anomaly detection methods from the perspective of graphs, there is a lack of attention to graph anomaly detection methods under the background of deep learning. In this paper, we first give the definitions of various kinds of anomalies in static graph and dynamic graph and investigate the deep neural network based graph representation learning method and its various applications in graph anomaly detection. Then we present the current situation of research on graph anomaly detection based on deep learning from the perspective of static graph and dynamic graph, and summarize the application scenarios and related data sets of graph anomaly detection. At last, we discuss the current challenges and future research directions of graph anomaly detection.

Key words: anomaly detection, deep learning, graph network, graph representation learning, graph neural network

中图分类号: