CS 691 - Team 5

1
CS 691 - Team 5
Biometric Authentication System

• Alex Wong
• Raheel Khan
• Rumeiz Hasseem
• Swati Bharati

2
Project Objectives

• Develop a biometric authentication system
• Application coded in Java
• Determine the feasibility of the Dichotomy Model
• Report results using standard authentication
  system performance statistics

Biometric Authentication System
CS 691 - Team 5
3
Dichotomy Model

• A statistically inferable approach to
  establishing the individuality of a biometric
• Classifies two biometric samples as coming either
  from the same person (intra-variation) or from
  two different people (inter-variation)
• Uses distance measure between two samples of the
  same class and between those of two different
  classes

Biometric Authentication System
CS 691 - Team 5
4
Objective of Dichotomy Model

• Validation of individuality of biometric data
  statistically
• Not the detection of differences of specific
  instances
• Find the individuality of the entire population
  based on the individuality of a sample of n
  people, where n is much less than the population.
• Allows inferential classification of individuals
  where large classes are involved and the whole
  population is not available for sampling

Biometric Authentication System
CS 691 - Team 5
5
Dichotomy vs. Polychotomy

• Binary decision, yes/no
• Authentication or Verification process
• A user is verified as being the person s/he
  claims to be
• More suitable for establishing individuality of a
  person, where number of classes is too large to
  completely sample, eg. population of an entire
  nation.

• One-of-many decision
• Identification process
• A user is identified from within a population of
  n users
• One-of-n response

Biometric Authentication System
CS 691 - Team 5
6
Original Feature Vector Data File
Biometric Authentication System
CS 691 - Team 5
7
Dichotomy Converted File
Biometric Authentication System
CS 691 - Team 5
8
Dichotomy Conversion Example

• First row
• SAME , 254 , 0.11431427822210534 - 0.0,..
• Fifth row
• DIFF, 254, 0.11431427822210534 -
  0.32848686484618,..
• Total number of
• Intra (SAME) class data samples
• m (m-1) n /2
• Inter (DIFF) class data samples
• m m n (n-1) /2
• Where
• n number of subjects
• m number samples from each subject
• For the given example
• Intra-class size 40 Inter-class size 150
  n4, m5

Biometric Authentication System
CS 691 - Team 5
9
Polychotomy to Dichotomy Conversion
Referencehttp//www.icgst.com/gvip/v5/P1150511001
.pdf 
Biometric Authentication System
CS 691 - Team 5
10
System Evaluation

• FRR (False Reject Rate)
• Same persons biometric data identified as coming
  from two different people
• FAR (False Accept Rate)
• Biometric data provided by two different people
  are classified as coming from the same person
• System Performance
• Biometric data correctly classified

Biometric Authentication System
CS 691 - Team 5
11
Project Specifications

• Convert training and testing files of n-class
  feature data into files of 2-class (inter and
  intra-class) dichotomy-model feature data
• Prepare sets of inter and intra-class data for
  training and testing
• Implement the nearest-neighbor technique to
  obtain accuracy results on the data (Euclidean
  distance)

Biometric Authentication System
CS 691 - Team 5
12
Application Design Decisions

• Allows for users to save Test Dichotomy Data both
  intra and inter class data sets
• Allows for users to also save the Train Dichotomy
  Data both intra and inter class data sets
• Users are able to view a log file of what action
  is currently being executed
• Results can be saved as a .html file to easily
  save and distribute them
• GUI is simple, clear and easy to use

Biometric Authentication System
CS 691 - Team 5
13
Application Demonstration
Biometric Authentication System
CS 691 - Team 5
14
CS 691 - Team 5
Biometric Authentication System Tutorial
15
Experimental Results

• Experiments Performed on data obtained from
• Mouse Movement biometric system
• Stylometry biometric system
• Keystroke biometric system
• Results show
• Overall System Performance
• FRR (False Reject Rate)
• FAR (False Accept Rate)

16
Mouse Movement Results Different subjects same
conditions

• Training set 115 samples from 5 subjects
• 30 samples each from 3 subjects, 15 samples from
  1 subject, 10 samples from 1 subject
• Testing set 90 samples from other 5 subjects
• 10 samples from 3 subjects, 30 samples each from
  2 subjects

Biometric Authentication System
CS 691 - Team 5
17
Mouse Movement ResultsUsing all subjects train
and test sets captured 3 weeks apart

• Training set 50 samples from all 5 subjects
• 10 samples from each 5 subjects
• Testing set 50 samples from all 5 subjects
• 10 samples from each 5 subjects approximately 3
  week interval

Biometric Authentication System
CS 691 - Team 5
18
Stylometry Results Different subjects same
conditions

• Training set 60 samples from 6 subjects
• 10 samples from each 6 subjects
• Testing set 60 samples from other 6 subjects
• 10 samples from each 6 subjects

Biometric Authentication System
CS 691 - Team 5
19
Stylometry ResultsTrain and test set on all
subjects by dividing the samples

• Training set 60 samples from all 12 subjects
• 5 samples from each 12 subjects
• Testing set 60 samples from all 12 subjects
• 5 samples from each 12 subjects

Biometric Authentication System
CS 691 - Team 5
20
Keystroke ResultsDifferent Subjects Same
Conditions

• Training set 90 samples from 18 subjects
• 5 samples from each 18 subjects
• Testing set 90 samples from other 18 subjects
• 5 samples from each 18 subjects all intra-inter
  data used

Biometric Authentication System
CS 691 - Team 5
21
Keystroke ResultsDifferent Subjects Same
Conditions Using a randomized set of 500
inter-class data

• Training set 90 samples from 18 subjects
• 5 samples from each 18 subjects
• Testing set 90 samples from other 18 subjects
• 5 samples from each 18 subjects 500 intra-inter
  sets used

Biometric Authentication System
CS 691 - Team 5
22
Keystroke ResultsTest results for old
keystroke data (180 samples 36 subjects 5
samples each) on same subjects and different
conditions.
Biometric Authentication System
CS 691 - Team 5
23
Keystroke ResultsLongitudinal authentication
test results on same subjects and conditions but
at two-week data collection interval.

• Training set (baseline) 20 samples from 4
  subjects
• 5 samples from each 4 subjects
• Testing set (2-week interval) 20 samples from 4
  subjects
• 5 samples from each 4 subjects

Biometric Authentication System
CS 691 - Team 5
24
Keystroke ResultsLongitudinal authentication
test results on same subjects and conditions but
at four-week data collection interval.

• Training set (baseline) 20 samples from 4
  subjects
• 5 samples from each 4 subjects
• Testing set (4-week interval) 20 samples from 4
  subjects
• 5 samples from each 4 subjects

Biometric Authentication System
CS 691 - Team 5
25
Project Achievements

• Utilized the dichotomy model in the
  authentication of biometric data obtained from
  the Keystroke, Stylometry and Mouse Movement
  biometric systems.
• Sought to establish that the dichotomy model is
  the preferred model over the polychotomy model
  when dealing with an enormous number of classes
  where the whole population is not available for
  sampling, that it is the statistically inferable
  approach.

Biometric Authentication System
CS 691 - Team 5
26
Summary of Results

• For the mouse movement and stylometry biometric
  data small number of users (classes)
• System performance between 66 and 76
• FAR and FRR high
• For the keystroke biometric data - large number
  of users (classes)
• System performance above 90 in most cases
• FAR less than 15 in most cases
• FRR almost always less than 10.

Biometric Authentication System
CS 691 - Team 5
27
Conclusion

• The results on the keystroke biometric data are
  encouraging and indicate that the dichotomy model
  may be a feasible solution to the authentication
  problem when a large number of classes are
  involved.

Biometric Authentication System
CS 691 - Team 5
28
Future Work

• Comparative analysis of the dichotomy
  authentication results with polychotomy
  authentication results obtained on the same
  keystroke biometric data.
• Study to see whether the results for the mouse
  movement and stylometry data improved
  significantly as the sample sizes increased.

Biometric Authentication System
CS 691 - Team 5
29
Please Visit Our Website
To obtain the latest downloads and information
please visit us online.
a
http//utopia.csis.pace.edu/cs691/2007-2008/team5/
index.html
Biometric Authentication System
CS 691 - Team 5
30
Thank you
Biometric Authentication System
CS 691 - Team 5