Open AccessA Study of Person Identification using Keystroke Dynamics and Statistical Analysis
Author(s) -
Nikhil Ashok Hegde
Publication year - 2018
Publication title -
international journal of engineering and management research
Language(s) - English
Resource type - Journals
eISSN - 2394-6962
pISSN - 2250-0758
DOI - 10.31033/ijemr.8.3.1
Subject(s) - keystroke dynamics , computer science , identification (biology) , keystroke logging , key (lock) , statistical analysis , set (abstract data type) , rank (graph theory) , data mining , feature (linguistics) , artificial intelligence , dwell time , standard deviation , machine learning , pattern recognition (psychology) , statistics , password , mathematics , computer security , s/key , medicine , clinical psychology , linguistics , philosophy , combinatorics , programming language , botany , biology
In this paper, a basic study of closed-setidentification using keystroke dynamics and simple statisticalanalysis has been carried out. Dwell time, flight time and oneadditional feature called key affinity are used as useridentifyingfeatures. The timing information is passedthrough a statistical layer to produce mean and standarddeviation. This information is combined with key affinity toidentify a rank-based person list. In conclusion, we comparethe performance of this setup with other setups. This workaims to suggest that a keystroke dynamics system relying onpure statistics as its underlying algorithm may not besufficiently accurate