F-FADE: Frequency Factorization for Anomaly Detection in Edge Streams

被引:25
作者
Chang, Yen-Yu [1 ]
Li, Pan [2 ]
Sosic, Rok [1 ]
Afifi, M. H. [3 ]
Schweighauser, Marco [3 ]
Leskovec, Jure [1 ]
机构
[1] Stanford Univ, Stanford, CA 94305 USA
[2] Purdue Univ, W Lafayette, IN 47907 USA
[3] Barracuda Networks, Campbell, CA USA
来源
WSDM '21: PROCEEDINGS OF THE 14TH ACM INTERNATIONAL CONFERENCE ON WEB SEARCH AND DATA MINING | 2021年
关键词
anomaly detection; dynamic network; account takeout protection;
D O I
10.1145/3437963.3441806
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Edge streams are commonly used to capture interactions in dynamic networks, such as email, social, or computer networks. The problem of detecting anomalies or rare events in edge streams has a wide range of applications. However, it presents many challenges due to lack of labels, a highly dynamic nature of interactions, and the entanglement of temporal and structural changes in the network. Current methods are limited in their ability to address the above challenges and to efficiently process a large number of interactions. Here, we propose F-FADE, a new approach for detection of anomalies in edge streams, which uses a novel frequency-factorization technique to efficiently model the time-evolving distributions of frequencies of interactions between node-pairs. The anomalies are then determined based on the likelihood of the observed frequency of each incoming interaction. F-FADE is able to handle in an online streaming setting a broad variety of anomalies with temporal and structural changes, while requiring only constant memory. Our experiments on one synthetic and six real-world dynamic networks show that F-FADE achieves state of the art performance and may detect anomalies that previous methods are unable to find.
引用
收藏
页码:589 / 597
页数:9
相关论文
共 49 条
[1]  
Aggarwal CC, 2011, PROC INT CONF DATA, P399, DOI 10.1109/ICDE.2011.5767885
[2]   Graph based anomaly detection and description: a survey [J].
Akoglu, Leman ;
Tong, Hanghang ;
Koutra, Danai .
DATA MINING AND KNOWLEDGE DISCOVERY, 2015, 29 (03) :626-688
[3]   RTM: Laws and a Recursive Generator for Weighted Time-Evolving Graphs [J].
Akoglu, Leman ;
McGlohon, Mary ;
Faloutsos, Christos .
ICDM 2008: EIGHTH IEEE INTERNATIONAL CONFERENCE ON DATA MINING, PROCEEDINGS, 2008, :701-706
[4]  
[Anonymous], 2010, P 16 ACM SIGKDD INT
[5]  
[Anonymous], 2016, P 11 INT C FUT INT T, DOI DOI 10.1145/2935663
[6]  
[Anonymous], 2012, SDM
[7]  
[Anonymous], 2015, P 2015 SIAM INT C DA
[8]  
[Anonymous], 2007, IN PROC KDD CUP WORK
[9]   Mining Persistent Activity in Continually Evolving Networks [J].
Belth, Caleb ;
Zheng, Xinyi ;
Koutra, Danai .
KDD '20: PROCEEDINGS OF THE 26TH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY & DATA MINING, 2020, :934-944
[10]  
[БЕНИН Андрей Владимирович BENIN Andrey V.], 2011, [Промышленное и гражданское строительство, Promyshlennoe i grazhdanskoe stroitel'stvo], P16