15 juin 2018
Ce document est lié à :
info:eu-repo/semantics/altIdentifier/doi/10.36824/2018-graf-ball
info:eu-repo/semantics/OpenAccess
Nicolas Ballier et al., « Investigating Keylogs as Time-Stamped Graphemics », HAL SHS (Sciences de l’Homme et de la Société), ID : 10.36824/2018-graf-ball
This article investigates keystroke data, in an attempt to articulate the microlevel of the graphemic level with the macrolevel of text structures. Analyz ing the timestamps of keylogs, we suggest a hierarchy of constituents inspired by speech data and focus on the interaction of graphemic structure, phonologi cal structure and textual structure within the dimension of time. We present the prototype of an R package designed to analyze keylog capture data, taking into account graphemic structures, syllable counts and parsing. Our R package un der development offers functions that can be used to analyze the various levels of graphemic constituents produced by typists, from syllable counts to ngram analysis.