Снижение размеров достаточной для обучения выборки за счет симметризации корреляционных связей биометрических данных / Иванов А. И., Ложников П. С., Серикова Ю. И. (2016)
Ukrainian

English  Cybernetics and Systems Analysis   /     Issue (2016, 52 (3))

Ivanov A.I., Lozhnikov P.S., Serikova Y.I.
Reducing the size of training-sufficient sampling due to symmetrization of correlation relationshps of biometric data

This paper shows that correlation coefficients obtained from small test samples for biometric data involve considerable uncertainty. This interferes with using them for machine training (setting) of classical quadratic forms and Bayesian networks. A method for symmetrizing correlation relationships is proposed. The requirement on the volume of biometric data is proved to be reduced considerably in this case. As a consequence, the setting (teaching) of quadratic forms and setting of maximum likelihood Bayesian networks become much more stable problems. This enables many-fold reduction in the requirement on the size of the training sample for an “own” biometric image. © 2016, Springer Science+Business Media New York.

Keywords: biometric identification, machine learning based on small test samples, symmetrization of correlation relationships, Artificial intelligence, Bayesian networks, Learning systems, Maximum likelihood, Number theory, Biometric identifications, Biometric image, Correlation coefficient, Machine trainings, Stable problem, symmetrization of correlation relationships, Test samples, Training sample, Biometrics


Cite:
Ivanov A.I., Lozhnikov P.S., Serikova Y.I. (2016). Reducing the size of training-sufficient sampling due to symmetrization of correlation relationshps of biometric data. Cybernetics and Systems Analysis, 52 (3), 49-56. doi: https://doi.org/10.1007/s10559-016-9838-x http://jnas.nbuv.gov.ua/article/UJRN-0000502503 [In Russian].


 

Інститут інформаційних технологій НБУВ


+38 (044) 525-36-24
Голосіївський просп., 3, к. 209
м. Київ, 03039, Україна