Evaluating the Performance of Biometric Identification Systems Using the Beta-binomial Distribution Model

Arnold Kiura Njuki^1, , Thomas Mageto¹ and Anthony Ngunyi²

¹Department of Statistics and Acturial Sciences, Jomo Kenyatta University of Agriculture and Technology, Juja, Kenya

²Department of Statistics and Acturial sciences, Dedan Kimathi University of Technology, Nyeri, Kenya

Cite this paper

Arnold Kiura Njuki, Thomas Mageto and Anthony Ngunyi. Evaluating the Performance of Biometric Identification Systems Using the Beta-binomial Distribution Model. American Journal of Applied Mathematics and Statistics. 2022; 10(1):14-21. doi: 10.12691/AJAMS-10-1-3

Abstract

Biometric authentication system has become a mainstream solution across industries and devices. From securing highly confidential data to unlocking smartphones, biometrics have eliminated the hassle of remembering multiple complex passwords and PINs. It means that nobody can gain access to a device or system without your presence. This paper discusses a method which could be used in the testing process of biometric systems on the side of users and customers. Large –scale biometric systems traditionally undergo a series of tests beyond technology and scenario testing. These large-scale system tests are typically at the system level, not just the biometric subsystem level, and occur multiple times in the life of a system in such forms as factory acceptance tests before shipment, site or system acceptance tests before initiating operations, and in- use tests to ensure that performance remains at acceptable levels and/or to reset thresholds or other technical parameters. The conventional statistical methods use the binomial distribution to estimate the expected number of failure, but in the field of the biometrics the probability parameter can’t be constant which means that it is necessary to describe a process. The results have shown that the probability is characterized with two parameters of the beta distribution, and these are predictable from a smaller sample of the investigated population with the maximum likelihood method.

Keywords

format, microsoft word template, style, insert, template

Copyright

This work is licensed under a Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

References

[1]	Rand R Wilcox, “Estimating the Parameters of the Beta- Binomial Distribution,” University of Southern California. (2016).
[2]	Gabor Á. Werner, László Hanka Ph.D, “Using the Beta-Binomial Distribution for the Analysis of Biometric Identification,” Óbuda University. (2015).
[3]	Dan Navarro, Amy Perfors, “An introduction to the Beta-Binomial model,” COMPSCI 3016: Computational Cognitive Science, University of Adelaide (2014).
[4]	László Hanka, “Mathematical Methods in Biometrics,” University of Óbuda, (2012).
[5]	In.van Tilborg H.C.A., Jajodia S. “Biometric Testing. In: (eds) Encyclopedia of Cryptography and Security.” Springer, Boston, MA. (2011).
[6]	J.Tong and D.Lord. “Beta-Binomial Models-CMRSC”. (2007).
[7]	Gammasi M, Lazzaroni M,Mishori M, Piuri V, Sana D, Scotti F. “Accuracy and performance of Biometric Systems”. (2004).
[8]	Dass SC, Zhu Y, Jain AK,Anal Mach Intell, “Validating a biometric authentication system: sample size requirements”. IEEE Trans Pattern (2006).
[9]	U.K. Biometrics Working Group. “Best practices in testing and reporting performance of biometric devices”, available in www.cesg.gov.uk/biometrics, (2000).
[10]	Michael E. Schuckers. “Using the Beta-binomial distribution to assess the performance of a biometic device” Submitted to Pattern Recognition (2003)[Online].
[11]	G. R. Doddington, W. Liggett, A. F. Martin, M. Przybocki, and D. A. Reynolds, “Sheep, Goats, Lambs and Wolves: A statistical analysis of speaker performance in the NIST speaker recognition evaluation”. (1998).
[12]	S. Silvey. “Statistical Inference” Halsted Press, New York. (1975).
[13]	Chatfield and Goodhart. “Applied Statistics” 19, 240-250 (1970).
[14]	Lin Hong, Yifei Wan, and Anil Jain. “Fingerprint Image Enhancement: Algorithm and Performance Evaluation” Pattern Recognition and Image Processing Laboratory Department of Computer Science, Michigan State University. 4-6 (1998).