Nonparametric self-exciting models for computer network traffic

File Description SizeFormat 
Price-Williams-Heard2019_Article_NonparametricSelf-excitingMode.pdfPublished version482.86 kBAdobe PDFView/Open
Title: Nonparametric self-exciting models for computer network traffic
Authors: Price-Williams, M
Heard, N
Item Type: Journal Article
Abstract: Connectivity patterns between nodes in a computer network can be interpreted and modelled as point processes where events in a process indicate connections being established for data to be sent along that edge. A model of normal connectivity behaviour can be constructed for each edge in a network by identifying key network user features such as seasonality or self-exciting behaviour, since events typically arise in bursts at particular times of day which may be peculiar to that edge. When monitoring a computer network in real time, unusual patterns of activity against the model of normality could indicate the presence of a malicious actor. A flexible, novel, nonparametric model for the excitation function of a Wold process is proposed for modelling the conditional intensities of network edges. This approach is shown to outperform standard seasonality and self-excitation models in predicting network connections, achieving well-calibrated predictions for event data collected from the computer networks of both Imperial College and Los Alamos National Laboratory.
Issue Date: 13-May-2019
Date of Acceptance: 17-Apr-2019
ISSN: 0960-3174
Publisher: Springer (part of Springer Nature)
Start Page: 1
End Page: 12
Journal / Book Title: Statistics and Computing
Copyright Statement: © The Author(s) 2019. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
Sponsor/Funder: GCHQ
Funder's Grant Number: Price-Williams PO-4177302
Keywords: Statistics & Probability
0104 Statistics
0802 Computation Theory and Mathematics
Publication Status: Published online
Open Access location:
Online Publication Date: 2019-05-13
Appears in Collections:Mathematics
Faculty of Natural Sciences

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commons