Publications

BibTex format

@article{Sanna:2021:10.1214/21-AOAS1540,
author = {Sanna, Passino F and Turcotte, MJM and Heard, NA},
doi = {10.1214/21-AOAS1540},
journal = {Annals of Applied Statistics},
pages = {1313--1332},
title = {Graph link prediction in computer networks using Poisson matrix  factorisation},
url = {http://dx.doi.org/10.1214/21-AOAS1540},
volume = {16},
year = {2021}
}

Download

RIS format (EndNote, RefMan)

TY  - JOUR
AB  - Graph link prediction is an important task in cyber-security: relationshipsbetween entities within a computer network, such as users interacting withcomputers, or system libraries and the corresponding processes that use them,can provide key insights into adversary behaviour. Poisson matrix factorisation(PMF) is a popular model for link prediction in large networks, particularlyuseful for its scalability. In this article, PMF is extended to includescenarios that are commonly encountered in cyber-security applications.Specifically, an extension is proposed to explicitly handle binary adjacencymatrices and include known covariates associated with the graph nodes. Aseasonal PMF model is also presented to handle dynamic networks. To allow themethods to scale to large graphs, variational methods are discussed forperforming fast inference. The results show an improved performance over thestandard PMF model and other common link prediction techniques.
AU  - Sanna,Passino F
AU  - Turcotte,MJM
AU  - Heard,NA
DO  - 10.1214/21-AOAS1540
EP  - 1332
PY  - 2021///
SN  - 1932-6157
SP  - 1313
TI  - Graph link prediction in computer networks using Poisson matrix  factorisation
T2  - Annals of Applied Statistics
UR  - http://dx.doi.org/10.1214/21-AOAS1540
UR  - http://hdl.handle.net/10044/1/89018
VL  - 16
ER  -

Download

Professor Nick Heard

Contact

Location

Summary

Citation

BibTex format

RIS format (EndNote, RefMan)