Download presentation
Presentation is loading. Please wait.
1
Keystroke Biometric Studies Keystroke Biometric Identification and Authentication on Long-Text Input Book chapter in Behavioral Biometrics for Human Identification (2009), edited by Liang Wang and Xin Geng Authors Charles C. Tappert, Mary Villani, and Sung-Hyuk Cha Summarizes keystroke biometric work 2005-2008 3 DPS dissertations 2 on identification, 1 on missing/incomplete data About 6 masters-level projects New material – authentication, longitudinal, touch-type model
2
Keystroke Biometric Studies Major Chapter Sections Introduction Keystroke Biometric System Experimental Design and Data Collection Experimental Results Conclusions and Future Work
3
Keystroke Biometric Studies Introduction Build a Case for Usefulness of Study Validate importance of study – applications Define keystroke biometric Appeal of keystroke over other biometrics Previous work on the keystroke biometric No direct study comparisons on same data Feature measurements Make case for using: data over the internet, long text input, free (arbitrary) text input Extends previous work by authors Summary of scope and methodology Summary of paper organization
4
Keystroke Biometric Studies Introduction Validate importance of study – applications Internet authentication application Authenticate (verify) student test-takers Internet identification application Identify perpetrators of inappropriate email Internet security for other applications Important as more businesses move toward e-commerce
5
Keystroke Biometric Studies Introduction Define Keystroke Biometric The keystroke biometric is one of the less-studied behavioral biometrics Based on the idea that typing patterns are unique to individuals and difficult to duplicate
6
Keystroke Biometric Studies Introduction Appeal of Keystroke Biometric Not intrusive – data captured as users type Users type frequently for business/pleasure Inexpensive – keyboards are common No special equipment necessary Can continue to check ID with keystrokes after initial authentication As users continue to type
7
Keystroke Biometric Studies Introduction Previous Work on Keystroke Biometric One early study goes back to typewriter input Identification versus authentication Most studies were on authentication Two commercial products on hardening passwords Few on identification (more difficult problem) Short versus long text input Most studies used short input – passwords, names Few used long text input –copy or free text Other keystroke problems studies One study detected fatigue, stress, etc. Another detected ID change via monitoring
8
Keystroke Biometric Studies Introduction No Direct Study Comparisons on Same Data No comparisons on a standard data set (desirable, available for many biometric and pattern recognition problems) Rather, researchers collect their own data Nevertheless, literature optimistic of keystroke biometric potential for security
9
Keystroke Biometric Studies Introduction Feature Measurements Features derived from raw data Key press times and key release times Each keystroke provides small amount of data Data varies from different keyboards, different conditions, and different entered texts Using long text input allows Use of good (statistical) feature measurements Generalization over keyboards, conditions, etc.
10
Keystroke Biometric Studies Introduction Make Case for Using Data over the internet Required by applications Long text input More and better features Higher accuracy Free text input Required by applications Predefined copy texts unacceptable
11
Keystroke Biometric Studies Introduction Extends Previous Work by Authors Previous keystroke identification study Ideal conditions Fixed text and Same keyboard for enrollment and testing Less ideal conditions Free text input Different keyboards for enrollment and testing
12
Keystroke Biometric Studies Introduction Summary of Scope and Methodology Determine distinctiveness of keystroke patterns Two application types Identification (1-of-n problem) Authentication (yes/no problem) Two indep. variables (4 data quadrants) Keyboard type – desktop versus laptop Entry mode – copy versus free text
13
Keystroke Biometric Studies Keystroke Biometric System Raw keystroke data capture Feature extraction Classification for identification Classification for authentication
14
Keystroke Biometric Studies Keystroke Biometric System Raw Keystroke Data Capture
15
Keystroke Biometric Studies Keystroke Biometric System Raw Keystroke Data Capture
16
Keystroke Biometric Studies Keystroke Biometric System Feature Extraction Mostly statistical features Averages and standard deviations Key press times Transition times between keystroke pairs Individual keys and groups of keys – hierarchy Percentage features Percentage use of non-letter keys Percentage use of mouse clicks Input rates – average time/keystroke
17
Keystroke Biometric Studies Keystroke Biometric System Feature Extraction A two-key sequence (th) showing the two transition measures
18
Keystroke Biometric Studies Keystroke Biometric System Feature Extraction Hierarchy tree for the 39 duration categories
19
Keystroke Biometric Studies Keystroke Biometric System Feature Extraction Hierarchy tree for the 35 transition categories
20
Keystroke Biometric Studies Keystroke Biometric System Feature Extraction Fallback procedure for few/missing samples When the number of samples is less than a fallback threshold, take the weighted average of the key’s mean and the fallback mean
21
Keystroke Biometric Studies Keystroke Biometric System Feature Extraction Two preprocessing steps Outlier removal Remove duration and transition times > threshold Feature standardization Convert features into the range 0-1
22
Keystroke Biometric Studies Keystroke Biometric System Classification for Identification Nearest neighbor using Euclidean distance Compare a test sample against the training samples, and the author of the nearest training sample is identified as the author of the test sample
23
Keystroke Biometric Studies Keystroke Biometric System Classification for Authentication Cha’s vector-distance (dichotomy) model
24
Keystroke Biometric Studies Experimental Design and Data Collection Design Two independent variables Keyboard type Desktop – all Dell Laptop – 90% Dell + IBM, Compaq, Apple, HP, Toshiba Input mode Copy task – predefined text Free text input – e.g., arbitrary email
25
Keystroke Biometric Studies Experimental Design and Data Collection Design
26
Keystroke Biometric Studies Experimental Design and Data Collection Data Collection Subjects provided samples in at least two quadrants Five samples per quadrant per subject Summary of subject demographics AgeFemaleMaleTotal Under 20151934 20-29122335 30-3951015 40-4971118 50+11516 All5068118
27
Keystroke Biometric Studies Experimental Results Identification experimental results Authentication experimental results Longitudinal study results System hierarchical model and parameters Hierarchical fallback model Outlier parameters Number of enrollment samples Input text length Probability distributions of statistical features
28
Keystroke Biometric Studies Experimental Results Identification Experimental Results Identification performance under ideal conditions (same keyboard type and input mode, leave-one-out procedure)
29
Keystroke Biometric Studies Experimental Results Identification Experimental Results Identification performance under non-ideal conditions (train on one file, test on another)
30
Keystroke Biometric Studies Experimental Design and Data Collection Design
31
Keystroke Biometric Studies Experimental Results Authentication Experimental Results Authentication performance under ideal conditions (weak enrollment: train on 18 subjects and test on 18 different subjects)
32
Keystroke Biometric Studies Experimental Results Longitudinal Study Results Identification – 13 subjects at 2-week intervals Average 6 arrow groups: 90% -> 85% -> 83% Authentication – 13 subjects at 2-week intervals Average 6 arrow groups: 90% -> 87% -> 85% Identification – 8 subjects at 2-year interval Average 6 arrow groups: 84% -> 67% Authentication – 8 subjects at 2-year interval Average 6 arrow groups: 94% -> 92% (all above results under non-ideal conditions)
33
Keystroke Biometric Studies Experimental Results System hierarchical model and parameters Touch-type hierarchy tree for durations
34
Keystroke Biometric Studies Experimental Results System hierarchical model and parameters Identification accuracy versus outlier removal passes
35
Keystroke Biometric Studies Experimental Results System hierarchical model and parameters Identification accuracy versus outlier removal distance (sigma)
36
Keystroke Biometric Studies Experimental Results System hierarchical model and parameters Identification accuracy versus enrollment samples
37
Keystroke Biometric Studies Experimental Results System hierarchical model and parameters Identification accuracy versus input text length
38
Keystroke Biometric Studies Experimental Results System hierarchical model and parameters Distributions of “u” duration times for each entry mode
39
Keystroke Biometric Studies Conclusions Results are important and timely as more people become involved in the applications of interest Authenticating online test-takers Identifying senders of inappropriate email High performance (accuracy) results if 2 or more enrollment samples/user Users use same keyboard type
40
Keystroke Biometric Studies Future Work Focus on user authentication Focus on Cha’s dichotomy model Develop strong/weak enrollment concepts Strong – system trained on actual users Weak – system trained on other (non-test) users Develop strategies to obtain ROC curves Run actual test-taker experiments
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.