Using Cluster of Algorithms for Biometric Authentication

The article presents the theoretical and practical aspects of using a multilayer perceptron. The choice of algorithms for solving the problems of biometric authentication by voice is carried out, their advantages and disadvantages are considered. It is proposed to combine a multilayer perceptron with selected algorithms into a cluster of algorithms. The advantages and disadvantages of the method of clustering algorithms are considered. A cluster of algorithms is being designed to carry out the authentication procedure. A biometric image of the user is offered. Models are trained based on each algorithm and the user's biometric image. An experiment is conducted with seven users and the results are analyzed.