Skip to main navigation Skip to search Skip to main content

A user authentication and identification model based on mouse dynamics

  • School of Computer Science and Technology (School of Software), Harbin Institute of Technology Weihai

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

Abstract

Most1 existing researches on identity authentication based on Mouse Dynamics require strictly restriction on experimental environment, and can only verify the legitimacy of the user without the ability to specifically recognize the identity of the impostor. To address this issue, we proposed an identification and authentication approach based on a weighted voting technology under an uncontrolled environment, which extracting both conventional statistical features and newly defined individual features of various operations. Two experiments are conducted with a dataset from 12 users. We obtained an accuracy of 84.1% in the identification experiment with an authentication time of 90s. In the authentication experiment, we get a FRR of 5.5% and a FAR of 8.8% with only 30s. Those noticeable results indicate the proposed method can be applied in practical environment.

Original languageEnglish
Title of host publicationProceedings of the 6th WASE International Conference on Information Engineering, ICIE 2017
PublisherAssociation for Computing Machinery
ISBN (Print)9781450352109
DOIs
StatePublished - 17 Aug 2017
Externally publishedYes
Event6th WASE International Conference on Information Engineering, ICIE 2017 - Dalian, Liaoning, China
Duration: 17 Aug 201718 Aug 2017

Publication series

NameACM International Conference Proceeding Series

Conference

Conference6th WASE International Conference on Information Engineering, ICIE 2017
Country/TerritoryChina
CityDalian, Liaoning
Period17/08/1718/08/17

Keywords

  • Authentication
  • Identification
  • Mouse dynamics
  • Voting technology

Fingerprint

Dive into the research topics of 'A user authentication and identification model based on mouse dynamics'. Together they form a unique fingerprint.

Cite this