Development of models for managing the population mass service system in a bank

Abstract

This paper presents models for evaluating the effectiveness of management tasks in Public Service Systems (PSSs), a mathematical model based on the analytical hierarchy process method for distributing elderly and conflicting customers along routes to a specific operator based on the emotional state and age of a customer. The models were trained and tested on the AffectNet and Adience datasets with four classes [angry, happy, neutral, sad] and [6-20, 25-30, 42-48, 60-98]. The accuracy of these models after training achieved in this study was compared with the results described in the scientific papers published in the Scopus Database. According to the experimental data obtained, the first and second developed models achieved validation accuracy of 76.30% and 75.1%, respectively.