Towards understanding behavioural biometric classifier performance over time and practice

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

2 Downloads (Pure)

Abstract

Behavioural biometrics in the context of security and authentication looks at the discriminative features of a user's measurable behaviour. This generally includes timing and location information related to, for example, screen touches and key presses, otherwise known as keystroke dynamics (KD). Research into KD has looked at discriminating features of behavioural patterns of expert typists, which are generally very stable, as well as patterns of novices, which are generally very unstable, if only because of rapid increases in skill level due to practice. The general population, however, at which such authentication solutions are aimed are not expert typists, and quickly move away from being novices, which we found causes significant degradation of biometric recognition performance over time.
This is because the biometric data entered at a later stage will increasingly differ from data gathered during enrollment. Accounting for practice effects in KD systems is diffcult, as not much is yet known about the way behaviour develops over time in biometrics literature. However, with the advent of open science, not only can we incorporate many new insights from the psychology of (motor) learning, but also re-analyse data sets gathered in this field in a biometric context.
In this paper we present initial analyses over a single data set, in which 36 (18
older, 18 young) participants were asked to complete a "password-entry"-like cognitive task, the Discrete Sequence Production task, a significant number of times spread over two sessions. Using out-of-the-box classiffiers, we found that
biometric recognition performance takes a long time, 75 repeats, to stabilize, as well as hints towards better initial biometric recognition performance for older participants than for younger.
Original languageEnglish
Title of host publicationProceedings of the 2017 Symposium on Information Theory and Signal Processing in the Benelux
Subtitle of host publicationMay 11-12, 2017, Delft University of Technology, Delft, the Netherlands
EditorsRichard Heusdens, Jos H. Weber
PublisherDelft University of Technology
Pages79-88
ISBN (Print)978-94-6186-811-4
Publication statusPublished - 2017
Event38th WIC Symposium on Information Theory in the Benelux 2017 - Delft, Netherlands
Duration: 11 May 201712 May 2017
Conference number: 38

Conference

Conference38th WIC Symposium on Information Theory in the Benelux 2017
CountryNetherlands
CityDelft
Period11/05/1712/05/17

Fingerprint

Biometrics
Classifiers
Authentication
Touch screens
Dynamical systems
Degradation

Cite this

Haasnoot, E., Barnhoorn, J. S., Spreeuwers, L. J., Veldhuis, R. N. J., & Verwey, W. B. (2017). Towards understanding behavioural biometric classifier performance over time and practice. In R. Heusdens, & J. H. Weber (Eds.), Proceedings of the 2017 Symposium on Information Theory and Signal Processing in the Benelux: May 11-12, 2017, Delft University of Technology, Delft, the Netherlands (pp. 79-88). Delft University of Technology.
Haasnoot, E. ; Barnhoorn, J.S. ; Spreeuwers, L.J. ; Veldhuis, R.N.J. ; Verwey, W.B. / Towards understanding behavioural biometric classifier performance over time and practice. Proceedings of the 2017 Symposium on Information Theory and Signal Processing in the Benelux: May 11-12, 2017, Delft University of Technology, Delft, the Netherlands. editor / Richard Heusdens ; Jos H. Weber. Delft University of Technology, 2017. pp. 79-88
@inproceedings{ec58e8126bb34a62b37fa0ff0ec509fd,
title = "Towards understanding behavioural biometric classifier performance over time and practice",
abstract = "Behavioural biometrics in the context of security and authentication looks at the discriminative features of a user's measurable behaviour. This generally includes timing and location information related to, for example, screen touches and key presses, otherwise known as keystroke dynamics (KD). Research into KD has looked at discriminating features of behavioural patterns of expert typists, which are generally very stable, as well as patterns of novices, which are generally very unstable, if only because of rapid increases in skill level due to practice. The general population, however, at which such authentication solutions are aimed are not expert typists, and quickly move away from being novices, which we found causes significant degradation of biometric recognition performance over time.This is because the biometric data entered at a later stage will increasingly differ from data gathered during enrollment. Accounting for practice effects in KD systems is diffcult, as not much is yet known about the way behaviour develops over time in biometrics literature. However, with the advent of open science, not only can we incorporate many new insights from the psychology of (motor) learning, but also re-analyse data sets gathered in this field in a biometric context.In this paper we present initial analyses over a single data set, in which 36 (18older, 18 young) participants were asked to complete a {"}password-entry{"}-like cognitive task, the Discrete Sequence Production task, a significant number of times spread over two sessions. Using out-of-the-box classiffiers, we found thatbiometric recognition performance takes a long time, 75 repeats, to stabilize, as well as hints towards better initial biometric recognition performance for older participants than for younger.",
author = "E. Haasnoot and J.S. Barnhoorn and L.J. Spreeuwers and R.N.J. Veldhuis and W.B. Verwey",
year = "2017",
language = "English",
isbn = "978-94-6186-811-4",
pages = "79--88",
editor = "Richard Heusdens and Weber, {Jos H.}",
booktitle = "Proceedings of the 2017 Symposium on Information Theory and Signal Processing in the Benelux",
publisher = "Delft University of Technology",
address = "Netherlands",

}

Haasnoot, E, Barnhoorn, JS, Spreeuwers, LJ, Veldhuis, RNJ & Verwey, WB 2017, Towards understanding behavioural biometric classifier performance over time and practice. in R Heusdens & JH Weber (eds), Proceedings of the 2017 Symposium on Information Theory and Signal Processing in the Benelux: May 11-12, 2017, Delft University of Technology, Delft, the Netherlands. Delft University of Technology, pp. 79-88, 38th WIC Symposium on Information Theory in the Benelux 2017, Delft, Netherlands, 11/05/17.

Towards understanding behavioural biometric classifier performance over time and practice. / Haasnoot, E.; Barnhoorn, J.S.; Spreeuwers, L.J.; Veldhuis, R.N.J.; Verwey, W.B.

Proceedings of the 2017 Symposium on Information Theory and Signal Processing in the Benelux: May 11-12, 2017, Delft University of Technology, Delft, the Netherlands. ed. / Richard Heusdens; Jos H. Weber. Delft University of Technology, 2017. p. 79-88.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Towards understanding behavioural biometric classifier performance over time and practice

AU - Haasnoot, E.

AU - Barnhoorn, J.S.

AU - Spreeuwers, L.J.

AU - Veldhuis, R.N.J.

AU - Verwey, W.B.

PY - 2017

Y1 - 2017

N2 - Behavioural biometrics in the context of security and authentication looks at the discriminative features of a user's measurable behaviour. This generally includes timing and location information related to, for example, screen touches and key presses, otherwise known as keystroke dynamics (KD). Research into KD has looked at discriminating features of behavioural patterns of expert typists, which are generally very stable, as well as patterns of novices, which are generally very unstable, if only because of rapid increases in skill level due to practice. The general population, however, at which such authentication solutions are aimed are not expert typists, and quickly move away from being novices, which we found causes significant degradation of biometric recognition performance over time.This is because the biometric data entered at a later stage will increasingly differ from data gathered during enrollment. Accounting for practice effects in KD systems is diffcult, as not much is yet known about the way behaviour develops over time in biometrics literature. However, with the advent of open science, not only can we incorporate many new insights from the psychology of (motor) learning, but also re-analyse data sets gathered in this field in a biometric context.In this paper we present initial analyses over a single data set, in which 36 (18older, 18 young) participants were asked to complete a "password-entry"-like cognitive task, the Discrete Sequence Production task, a significant number of times spread over two sessions. Using out-of-the-box classiffiers, we found thatbiometric recognition performance takes a long time, 75 repeats, to stabilize, as well as hints towards better initial biometric recognition performance for older participants than for younger.

AB - Behavioural biometrics in the context of security and authentication looks at the discriminative features of a user's measurable behaviour. This generally includes timing and location information related to, for example, screen touches and key presses, otherwise known as keystroke dynamics (KD). Research into KD has looked at discriminating features of behavioural patterns of expert typists, which are generally very stable, as well as patterns of novices, which are generally very unstable, if only because of rapid increases in skill level due to practice. The general population, however, at which such authentication solutions are aimed are not expert typists, and quickly move away from being novices, which we found causes significant degradation of biometric recognition performance over time.This is because the biometric data entered at a later stage will increasingly differ from data gathered during enrollment. Accounting for practice effects in KD systems is diffcult, as not much is yet known about the way behaviour develops over time in biometrics literature. However, with the advent of open science, not only can we incorporate many new insights from the psychology of (motor) learning, but also re-analyse data sets gathered in this field in a biometric context.In this paper we present initial analyses over a single data set, in which 36 (18older, 18 young) participants were asked to complete a "password-entry"-like cognitive task, the Discrete Sequence Production task, a significant number of times spread over two sessions. Using out-of-the-box classiffiers, we found thatbiometric recognition performance takes a long time, 75 repeats, to stabilize, as well as hints towards better initial biometric recognition performance for older participants than for younger.

M3 - Conference contribution

SN - 978-94-6186-811-4

SP - 79

EP - 88

BT - Proceedings of the 2017 Symposium on Information Theory and Signal Processing in the Benelux

A2 - Heusdens, Richard

A2 - Weber, Jos H.

PB - Delft University of Technology

ER -

Haasnoot E, Barnhoorn JS, Spreeuwers LJ, Veldhuis RNJ, Verwey WB. Towards understanding behavioural biometric classifier performance over time and practice. In Heusdens R, Weber JH, editors, Proceedings of the 2017 Symposium on Information Theory and Signal Processing in the Benelux: May 11-12, 2017, Delft University of Technology, Delft, the Netherlands. Delft University of Technology. 2017. p. 79-88