107
User Authentication
6.4 Evaluation of Behavioral Biometric Techniques
Jain [20] presents an extensive comparison of various behavioral biometric techniques in
Table 6.1. He proposes a table that presents evaluation of those parameters for behavioral
biometric technique in the following scale: H = high, M = median, L = low.
The following parameters are generally used to evaluate the biometric technique.
6.4.1 Universality
It defines how commonly biometric is found individually. The number of users of
computer-/handheld-device-enabled Internet connection is rapidly increasing and
the keyboard has become a mandatory device. But entered texts may not have the
same sequence for all the users. Universality of keystroke dynamics is the ratio
between the number of users having same sequence of characters and total number of
users [19].
6.4.2 Uniqueness
It is a measurement of how well biometric separates individuals from others. It is globally
accepted that typing rhythm is unique for all. Most of the time, experiments were done in
a controlled environment with impressive results. But due to a large number of affecting
factors, it is not accepted in practice.
6.4.3 Permanence
It defines that how much complexity to measure in acquisition. Timing parameters are
very easy to calculate in keystroke dynamics. But to measure key pressure and acceleration, extra sensors are required. Nowadays, advance sensing device and accelerometer are
embedded in each smart phone to measure pressure and acceleration.
6.4.4 Collectability
It represents how well it resists the aging problem. Typing behavior varies during the
course of a day or between 2 days. So updating the mechanism is necessary.
6.4.5 Performance
Performance of all the behavioral biometrics is low. Many papers have been published
with impressive results in a controlled environment. But in practice, it is very hard to
achieve impressive results.
6.4.6 Acceptability
It describes the degree of approval of a technology. Many commercial products have
been introduced; trademarks of the products are BioPassword, AuthenWare, TypeSence,
Phylock, etc.
User Authentication
6.4 Evaluation of Behavioral Biometric Techniques
Jain [20] presents an extensive comparison of various behavioral biometric techniques in
Table 6.1. He proposes a table that presents evaluation of those parameters for behavioral
biometric technique in the following scale: H = high, M = median, L = low.
The following parameters are generally used to evaluate the biometric technique.
6.4.1 Universality
It defines how commonly biometric is found individually. The number of users of
computer-/handheld-device-enabled Internet connection is rapidly increasing and
the keyboard has become a mandatory device. But entered texts may not have the
same sequence for all the users. Universality of keystroke dynamics is the ratio
between the number of users having same sequence of characters and total number of
users [19].
6.4.2 Uniqueness
It is a measurement of how well biometric separates individuals from others. It is globally
accepted that typing rhythm is unique for all. Most of the time, experiments were done in
a controlled environment with impressive results. But due to a large number of affecting
factors, it is not accepted in practice.
6.4.3 Permanence
It defines that how much complexity to measure in acquisition. Timing parameters are
very easy to calculate in keystroke dynamics. But to measure key pressure and acceleration, extra sensors are required. Nowadays, advance sensing device and accelerometer are
embedded in each smart phone to measure pressure and acceleration.
6.4.4 Collectability
It represents how well it resists the aging problem. Typing behavior varies during the
course of a day or between 2 days. So updating the mechanism is necessary.
6.4.5 Performance
Performance of all the behavioral biometrics is low. Many papers have been published
with impressive results in a controlled environment. But in practice, it is very hard to
achieve impressive results.
6.4.6 Acceptability
It describes the degree of approval of a technology. Many commercial products have
been introduced; trademarks of the products are BioPassword, AuthenWare, TypeSence,
Phylock, etc.