| Abstrakt: |
The article offers an approach to assessing customer satisfaction based on the recognition of emotional states of clients before and after the provision of services. A developed method of assessing the quality of customer service, based on recognition of emotions, is given. As a tool, the method of artificial immune systems is used to recognize emotions by facial expressions in images. In the work, using one of the methods of artificial immune systems, it was possible to achieve a maximum accuracy of 80% for the task of recognizing the basic emotions defined by Paul Ekman. These values were achieved on the Cohn-Kanade+ dataset. To build such a system, the most popular approaches to recognizing emotions in images were considered, as well as key concepts of typing of emotions. The proposed model used an approach based on computer vision using facial markings. The obtained coordinates of the points of the face were converted into real features, and then, their number was reduced using the method of principal components. An automated system for assessing the quality of customer service on the basis of the emotional satisfaction of clients has been developed, making it possible to take into account the emotional states of clients after the service delivery process. [ABSTRACT FROM AUTHOR] |
Full text from SpringerLink