Analysis of the possibilities of using biometric characteristics to identify a potential internal violator based on his psycho-emotional state

The main biometric characteristics reflecting changes in the psycho-emotional state of the user of the information system are considered. Their ranking was performed using the method of paired comparisons, as a result of which the voice and keyboard handwriting were identified as the most suitable for further research. The criteria for preliminary identification of potential internal information security violators based on changes in the considered biometric characteristics are defined. A convolutional neural network model has been developed and tested to solve this problem.