Dissertations, Theses, and Capstone Projects
Date of Degree
6-2016
Document Type
Thesis
Degree Name
M.A.
Program
Linguistics
Advisor
Andrew Rosenberg
Advisor
Martin Chodorow
Subject Categories
Computational Linguistics
Keywords
computational linguistics, biometrics, keystroke dynamics, cognitive science
Abstract
The process of producing written text is complex and constrained by pressures that range from physical to psychological. In a series of three sets of experiments, this thesis demonstrates the effects of linguistic context on the timing patterns of the production of keystrokes. We elucidate the effect of linguistic context at three different levels of granularity: The first set of experiments illustrate how the nontraditional syntax of a single linguistic construct, the multi-word expression, can create significant changes in keystroke production patterns. This set of experiments is followed by a set of experiments that test the hypothesis on the entire linguistic output of an individual. By taking into account linguistic context, we are able to create more informative feature-sets, and utilize these to improve the accuracy of keystroke dynamic-based user authentication. Finally, we extend our findings to entire populations, or demographic cohorts. We show that typing patterns can be used to predict a group's gender, native language and dominant hand. In addition, keystroke patterns can shed light on the cognitive complexity of a task that a typist is engaged in. The findings of these experiments have far-reaching implications for linguists, cognitive scientists, computer security researchers and social scientists.
Recommended Citation
Goodkind, Adam, "Utilizing Linguistic Context To Improve Individual and Cohort Identification in Typed Text" (2016). CUNY Academic Works.
https://academicworks.cuny.edu/gc_etds/1360