Identification of a Smartphone User via Keystroke Analysis

Abstract

Keystroke analysis has been an accepted method for user identification and authentication since the early 1980s. Most of the research in this field of biometrics has focused on traditional computer keyboards, with very few experiments performed on touchscreen keyboards found on modern smartphones. This study focused on identifying a smartphone user based on typing samples input by copying pre-written text, as well as spontaneously-authored free text. Features used for identification were duration of key press, as well as bigram and trigram transitions. User classification based on duration features proved to be successful in 70 percent of inputs to our k-nearest neighbors classifier.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 2014
Accession Number
ADA607855

Entities

People

  • Samuel B. Fleming

Organizations

  • Naval Postgraduate School

Tags

Communities of Interest

  • Cyber
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Authentication
  • Biometric Security
  • Biometrics
  • Classification
  • Computer Science
  • Computers
  • Data Mining
  • Identification
  • Information Science
  • Machine Learning
  • Mobile Devices
  • Mobile Operating Systems
  • Mobile Phones
  • Network Science
  • Smartphones
  • Supervised Machine Learning

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Computer Science/Computer Engineering/Data Science/Digital Signal Processing.
  • Neural Network Machine Learning.