###
Journal of Software:2020.31(3):748-762

基于图神经网络的动态网络异常检测算法
郭嘉琰,李荣华,张岩,王国仁
(北京大学 信息科学技术学院, 北京 100871;北京理工大学 计算机学院, 北京 100081)
Graph Neural Network Based Anomaly Detection in Dynamic Networks
GUO Jia-Yan,LI Rong-Hua,ZHANG Yan,WANG Guo-Ren
(School of Electronics Engineering and Computer Science, Peking University, Beijing 100871, China;School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081, China)
Abstract
Chart / table
Reference
Similar Articles
Article :Browse 1343   Download 1862
Received:July 19, 2019    Revised:November 25, 2019
> 中文摘要: 动态变化的图数据在现实应用中广泛存在,有效地对动态网络异常数据进行挖掘,具有重要的科学价值和实践意义.大多数传统的动态网络异常检测算法主要关注于网络结构的异常,而忽视了节点和边的属性以及网络变化的作用.提出一种基于图神经网络的异常检测算法,将图结构、属性以及动态变化的信息引入模型中,来学习进行异常检测的表示向量.具体地,改进图上无监督的图神经网络框架DGI,提出一种面向动态网络无监督表示学习算法Dynamic-DGI.该方法能够同时提取网络本身的异常特性以及网络变化的异常特性,用于表示向量的学习.实验结果表明,使用该算法学得的网络表示向量进行异常检测,得到的结果优于最新的子图异常检测算法SpotLight,并且显著优于传统的网络表示学习算法.除了能够提升异常检测的准确度,该算法也能够挖掘网络中存在的有实际意义的异常.
Abstract:Dynamic graph structured data is ubiquitous in real-life applications. Mining outliers on dynamic networks is an important problem, which is very useful for many practical applications. Most traditional network outlier detection algorithms focus mainly on the strutraulal anomaly, ignoring the nodes and edges' attributes, and the time-varying features as well. This study proposes a graph neural network based network anomaly detection algorithm which can capture the nodes and edges' attributes and time-varying features and fully uses these features to learn a representation vector for each node. Specifically, the proposed algorithm improves an unsupervised graph neural network framework called DGI. Based on DGI, a new danamic DGI algorithm is proposed, which is called Dynamic-DGI, for dynamic networks. Dynamic-DGI can simultaneously extracts the abnormal characteristics of the network itself and the abnormal characteristics of the network changes. The experimental results show that the proposed algorithm is better than the state-of-the-art anomaly detection algorithm SpotLight, and is significantly better than the traditional network representation learning algorithms. In addition to improving the accuracy, the proposed algorithmis also able to mine interesting anomalies in the network.
文章编号:     中图分类号:    文献标志码:
基金项目:国家自然科学基金(61772346,U1809206,61532001,61332006,61332014,61328202,U1401256);教育部-中国移动科研基金(MCM20170503) 国家自然科学基金(61772346,U1809206,61532001,61332006,61332014,61328202,U1401256);教育部-中国移动科研基金(MCM20170503)
Foundation items:National Natural Science Foundation of China (61772346, U1809206, 61532001, 61332006, 61332014, 61328202, U1401256); China MOE and China Mobile Joint Research Foundation (MCM20170503)
Reference text:

郭嘉琰,李荣华,张岩,王国仁.基于图神经网络的动态网络异常检测算法.软件学报,2020,31(3):748-762

GUO Jia-Yan,LI Rong-Hua,ZHANG Yan,WANG Guo-Ren.Graph Neural Network Based Anomaly Detection in Dynamic Networks.Journal of Software,2020,31(3):748-762