Keystroke Biometric : ROC Experiments Team Abhishek Kanchan Priyanka Ranadive Sagar Desai Pooja Malhotra Ning Wang
Keystroke Biometric : ROC Experiments WHAT IS KEYSTROKE BIOMETRIC ? The keystroke biometric is one of the less- studied behavioral biometrics. Keystroke biometric systems measure typing characteristics believed to be unique to an individual and difficult to duplicate. Used for Identification Used for Authentication Developed over the past 6+ years
Keystroke Biometric : ROC Experiments Introduction to ROC Curves Used for binary decisions Signal detection – signal / no signal Medical diagnosis – disease / no disease Biometric authentication – you are the person you claim to be / you are not
Keystroke Biometric : ROC Experiments Introduction to ROC Curves In biometrics the ROC curve varies from FAR=1 & FRR=0 at one end to FAR=0 & FRR=1 at other FAR = False Accept Rate – the rate an imposter is falsely accepted FRR = False Reject Rate – the rate the correct person is falsely rejected ROC Charts are expressed in terms of percentages (0- 100%) or probabilities (0-1). These are used interchangeably.
Keystroke Biometric : ROC Experiments ROC Authentication Analogy Supreme Court – nine judges – Usual procedure – majority required to make decision – Like 9NN needing majority to authenticate a user ROC Curve – effectively creates many potential procedures and provides FAR/FRR tradeoff for each (here is the m-kNN method) – Need 9 votes to make decision (very conservative) – Need 8, 7, 6 votes to make decision (conservative) – Need 5 votes to make decision (majority) – Need 4, 3, 2 votes to make decision (liberal) – Need 1 or even 0 votes to make decision (very liberal)
Keystroke Biometric : ROC Experiments ROC EXPERIMENTS Derived from four nonparametric techniques. ‘Weak' and ‘Strong' training experiments. – Weak Enrollment data, only non-test- subject data is used to train the system. – Strong enrollment uses test-subject data to train the system, and then uses independent (different) test-subject data to test the system. Large Data Experiments
Keystroke Biometric : ROC Experiments SYSTEM OVERVIEW
Keystroke Biometric : ROC Experiments Parametric Procedures Parametric techniques are well studied. Data follows a normal or Gaussian distribution. Vary a threshold to obtain the tradeoff between FAR/FRR. Probability density functions can be calculated without estimation. Parametric ROC - Probability Density Function - Adapted from Cha, et al (2009)
Keystroke Biometric : ROC Experiments Cha Dichotomy Model Simplifies complexity Transforms a feature space into a distance vector space. Uses distance measures. Multi-class to two Class Transformation Process, Adapted from Yoon et al (2005)
Keystroke Biometric : ROC Experiments Pure Rank Method – m-kNN Pure Rank Method. Evaluate the top 7 NN. Q is authenticated if # within-class matches is >= decision threshold of 4NN. Unweighted. All W’s are equal in weight.
Keystroke Biometric : ROC Experiments Rank Method Weighted by Rank Order wm-kNN Authenticate if W choices are > weighted match (m) Score varies from 0 to =k(k+1)/2 For every m, FAR/FRR pair or ROC point. If m=0, FAR=1, FAR=0 …All users accepted. If m=15, FAR=small, FRR=large, few Q’s accepted.
Keystroke Biometric : ROC Experiments m-kNN and wm-kNN ROC’s LapFree – Weak Training
Keystroke Biometric : ROC Experiments Distance Threshold Method t-kNN A positive vote is within a distance threshold from the user’s sample. Uses feature vector space distances only. At 0, no distance vectors are authenticated. FAR=0, FRR=100%. At t=100, all distance vectors are authenticated. FAR=100, FRR=0.
Keystroke Biometric : ROC Experiments Threshold (t-kNN) Method DeskFree (left) and LapFree (right) Data
Keystroke Biometric : ROC Experiments Threshold (ht-kNN) Method Weighted vote based on distances to the kNN. Hybrid of rank method and vector space distances. For each test sample, the within- class weight (WCW) is calculated based on the distance vectors. DeskFree (left) and LapFree (right) Data
Keystroke Biometric : ROC Experiments Weak & Strong Training
Keystroke Biometric : ROC Experiments DELIVERABLE Deliverable 5 – Authentication Experiments – Ideal Conditions/ Weak Enrollment Part I Status – Completed Deliverable 6 - Authentication Experiments – Ideal Conditions/ Weak Enrollment Part II Status – Completed Deliverable 7 – Enhance and Correct Refactor-BAS.jar ROC interface Status - Completed
Keystroke Biometric : ROC Experiments DELIVERABLE 7 Implement Perl ROC with threshold logic in JAVA. Unify the code in Java which was supported by a Perl program earlier for calculating ROC threshold Values. Maintain the performance of Perl code in Java. Some changes in User Interface of ROC program.
Keystroke Biometric : ROC Experiments UI CHANGES
Keystroke Biometric : ROC Experiments
TEAM COMMUNICATION Google Group for information sharing and discussion Skype Meetings s Personal Meetings Documented Minutes of Meeting Team Website status updates Assigned Task progress check by team leader
Keystroke Biometric : ROC Experiments Questions?