Continuous user authentication via unlabeled phone movement patterns

Rajesh Kumar, Partha Pratim Kundu, Diksha Shukla, Vir Phoha

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

In this paper, we propose a novel continuous authentication system for smartphone users. The proposed system entirely relies on unlabeled phone movement patterns collected through smartphone accelerometer. The data was collected in a completely unconstrained environment over five to twelve days. The contexts of phone usage were identified using k-means clustering. Multiple profiles, one for each context, were created for every user. Five machine learning algorithms were employed for classification of genuine and impostors. The performance of the system was evaluated over a diverse population of 57 users. The mean equal error rates achieved by Logistic Regression, Neural Network, kNN, SVM, and Random Forest were 13.7%, 13.5%, 12.1%, 10.7%, and 5.6% respectively. A series of statistical tests were conducted to compare the performance of the classifiers. The suitability of the proposed system for different types of users was also investigated using the failure to enroll policy.

Original languageEnglish (US)
Title of host publicationIEEE International Joint Conference on Biometrics, IJCB 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages177-184
Number of pages8
Volume2018-January
ISBN (Electronic)9781538611241
DOIs
StatePublished - Jan 29 2018
Event2017 IEEE International Joint Conference on Biometrics, IJCB 2017 - Denver, United States
Duration: Oct 1 2017Oct 4 2017

Other

Other2017 IEEE International Joint Conference on Biometrics, IJCB 2017
CountryUnited States
CityDenver
Period10/1/1710/4/17

Fingerprint

Smartphones
Authentication
statistical tests
machine learning
Statistical tests
logistics
accelerometers
classifiers
Accelerometers
Learning algorithms
Learning systems
Logistics
regression analysis
Classifiers
Neural networks
profiles

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Instrumentation
  • Signal Processing
  • Biomedical Engineering

Cite this

Kumar, R., Kundu, P. P., Shukla, D., & Phoha, V. (2018). Continuous user authentication via unlabeled phone movement patterns. In IEEE International Joint Conference on Biometrics, IJCB 2017 (Vol. 2018-January, pp. 177-184). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BTAS.2017.8272696

Continuous user authentication via unlabeled phone movement patterns. / Kumar, Rajesh; Kundu, Partha Pratim; Shukla, Diksha; Phoha, Vir.

IEEE International Joint Conference on Biometrics, IJCB 2017. Vol. 2018-January Institute of Electrical and Electronics Engineers Inc., 2018. p. 177-184.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kumar, R, Kundu, PP, Shukla, D & Phoha, V 2018, Continuous user authentication via unlabeled phone movement patterns. in IEEE International Joint Conference on Biometrics, IJCB 2017. vol. 2018-January, Institute of Electrical and Electronics Engineers Inc., pp. 177-184, 2017 IEEE International Joint Conference on Biometrics, IJCB 2017, Denver, United States, 10/1/17. https://doi.org/10.1109/BTAS.2017.8272696
Kumar R, Kundu PP, Shukla D, Phoha V. Continuous user authentication via unlabeled phone movement patterns. In IEEE International Joint Conference on Biometrics, IJCB 2017. Vol. 2018-January. Institute of Electrical and Electronics Engineers Inc. 2018. p. 177-184 https://doi.org/10.1109/BTAS.2017.8272696
Kumar, Rajesh ; Kundu, Partha Pratim ; Shukla, Diksha ; Phoha, Vir. / Continuous user authentication via unlabeled phone movement patterns. IEEE International Joint Conference on Biometrics, IJCB 2017. Vol. 2018-January Institute of Electrical and Electronics Engineers Inc., 2018. pp. 177-184
@inproceedings{85a7851d94134048880e060e4e20b48c,
title = "Continuous user authentication via unlabeled phone movement patterns",
abstract = "In this paper, we propose a novel continuous authentication system for smartphone users. The proposed system entirely relies on unlabeled phone movement patterns collected through smartphone accelerometer. The data was collected in a completely unconstrained environment over five to twelve days. The contexts of phone usage were identified using k-means clustering. Multiple profiles, one for each context, were created for every user. Five machine learning algorithms were employed for classification of genuine and impostors. The performance of the system was evaluated over a diverse population of 57 users. The mean equal error rates achieved by Logistic Regression, Neural Network, kNN, SVM, and Random Forest were 13.7{\%}, 13.5{\%}, 12.1{\%}, 10.7{\%}, and 5.6{\%} respectively. A series of statistical tests were conducted to compare the performance of the classifiers. The suitability of the proposed system for different types of users was also investigated using the failure to enroll policy.",
author = "Rajesh Kumar and Kundu, {Partha Pratim} and Diksha Shukla and Vir Phoha",
year = "2018",
month = "1",
day = "29",
doi = "10.1109/BTAS.2017.8272696",
language = "English (US)",
volume = "2018-January",
pages = "177--184",
booktitle = "IEEE International Joint Conference on Biometrics, IJCB 2017",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Continuous user authentication via unlabeled phone movement patterns

AU - Kumar, Rajesh

AU - Kundu, Partha Pratim

AU - Shukla, Diksha

AU - Phoha, Vir

PY - 2018/1/29

Y1 - 2018/1/29

N2 - In this paper, we propose a novel continuous authentication system for smartphone users. The proposed system entirely relies on unlabeled phone movement patterns collected through smartphone accelerometer. The data was collected in a completely unconstrained environment over five to twelve days. The contexts of phone usage were identified using k-means clustering. Multiple profiles, one for each context, were created for every user. Five machine learning algorithms were employed for classification of genuine and impostors. The performance of the system was evaluated over a diverse population of 57 users. The mean equal error rates achieved by Logistic Regression, Neural Network, kNN, SVM, and Random Forest were 13.7%, 13.5%, 12.1%, 10.7%, and 5.6% respectively. A series of statistical tests were conducted to compare the performance of the classifiers. The suitability of the proposed system for different types of users was also investigated using the failure to enroll policy.

AB - In this paper, we propose a novel continuous authentication system for smartphone users. The proposed system entirely relies on unlabeled phone movement patterns collected through smartphone accelerometer. The data was collected in a completely unconstrained environment over five to twelve days. The contexts of phone usage were identified using k-means clustering. Multiple profiles, one for each context, were created for every user. Five machine learning algorithms were employed for classification of genuine and impostors. The performance of the system was evaluated over a diverse population of 57 users. The mean equal error rates achieved by Logistic Regression, Neural Network, kNN, SVM, and Random Forest were 13.7%, 13.5%, 12.1%, 10.7%, and 5.6% respectively. A series of statistical tests were conducted to compare the performance of the classifiers. The suitability of the proposed system for different types of users was also investigated using the failure to enroll policy.

UR - http://www.scopus.com/inward/record.url?scp=85046285347&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046285347&partnerID=8YFLogxK

U2 - 10.1109/BTAS.2017.8272696

DO - 10.1109/BTAS.2017.8272696

M3 - Conference contribution

VL - 2018-January

SP - 177

EP - 184

BT - IEEE International Joint Conference on Biometrics, IJCB 2017

PB - Institute of Electrical and Electronics Engineers Inc.

ER -