Insights from curve fitting models in mouse dynamics authentication systems

Yi Xiang Marcus Tan, Alexander Binder, Arunava Roy

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

5 Citations (Scopus)

Abstract

In this work, we investigate the impact of various curve smoothing approaches on the performance of prediction tasks using mouse movement sequences as inputs in the setting of user authentication. We make use of three time-series forecasting modelling techniques as alternatives and we perform a comparison. We demonstrate our findings for the Balabit challenge mouse movement dataset.

Original languageEnglish
Title of host publication2017 IEEE Conference on Applications, Information and Network Security, AINS 2017
Place of PublicationPiscataway NJ USA
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages42-47
Number of pages6
ISBN (Electronic)9781538607251
DOIs
Publication statusPublished - 2017
Externally publishedYes
EventIEEE Conference on Applications, Information and Network Security 2017 - Miri, Sarawak, Malaysia
Duration: 13 Nov 201714 Nov 2017
https://ieeexplore.ieee.org/xpl/conhome/8262680/proceeding (Proceedings)
https://www.aconf.org/conf_108761.2017_IEEE_Conference_on_Application,_Information_and_Network_Security.html (Website)

Conference

ConferenceIEEE Conference on Applications, Information and Network Security 2017
Abbreviated titleAINS 2017
CountryMalaysia
CityMiri, Sarawak
Period13/11/1714/11/17
Internet address

Keywords

  • Active Authentication
  • Cubic Splines
  • Linear Support Vector Machines
  • Mouse Dynamics
  • Time Series Forecasting Models

Cite this