Learning from Smartphone Location Data as Anomaly Detection for Behavioral Authentication through Deep Neuroevolution

Mhd Irvan, Tran Thao, Ryosuke Kobayashi, Toshiyuki Nakata, Rie Yamaguchi

Abstract

Passwords and face recognition are some examples of many approaches to authenticate smartphone users. These approaches typically authenticate users at an initial log-in or unlock session, and there are risks of an unauthorized person using the authenticated account if the smartphone owner lose their device while still in unlocked status. Because of this reason, there is a necessity to continuously authenticate from time to time. Passwords and biological biometrics-based authentication procedures are impractical for this kind of situation because they require constant interruption. In this early research we are applying a behavioral authentication approach implementing location history data to implicitly authenticate users. Traits derived from users’ movements are easy to monitor and hard to fake. Previously visited locations represent patterns within people’s daily behaviors and in this paper we are proposing deep learning method evolved by genetic algorithms to recognize such patterns and to correctly authenticate people that match the patterns.

Download


Paper Citation


in Harvard Style

Irvan M., Thao T., Kobayashi R., Nakata T. and Yamaguchi R. (2021). Learning from Smartphone Location Data as Anomaly Detection for Behavioral Authentication through Deep Neuroevolution.In Proceedings of the 7th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP, ISBN 978-989-758-491-6, pages 723-728. DOI: 10.5220/0010395407230728


in Bibtex Style

@conference{icissp21,
author={Mhd Irvan and Tran Thao and Ryosuke Kobayashi and Toshiyuki Nakata and Rie Yamaguchi},
title={Learning from Smartphone Location Data as Anomaly Detection for Behavioral Authentication through Deep Neuroevolution},
booktitle={Proceedings of the 7th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP,},
year={2021},
pages={723-728},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010395407230728},
isbn={978-989-758-491-6},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 7th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP,
TI - Learning from Smartphone Location Data as Anomaly Detection for Behavioral Authentication through Deep Neuroevolution
SN - 978-989-758-491-6
AU - Irvan M.
AU - Thao T.
AU - Kobayashi R.
AU - Nakata T.
AU - Yamaguchi R.
PY - 2021
SP - 723
EP - 728
DO - 10.5220/0010395407230728