SOFTWARE USER ENGAGEMENT: METHODS OF EVALUATION AND APPLICATIONS

Abstract

The article explores phenomenon of engagement as a socio-psychological category that determines physical, emotional and intellectual state of a person involved in any activity. Further, engagement is considered as one of the factors that significantly affects the effectiveness and success of a person's professional activity. Methods for assessing engagement, distinguished by the types of data collected and approaches for their collection, processing and interpretation in the context of engagement, are analyzed. Existing software tools for assessing engagement based on self-assessment and systems for automatic assessment of engagement based on computer vision and machine learning methods are described. The concept of software user engagement as a user's reaction to interaction with this software is considered. The possibilities of using the results of user engagement assessment to assess the activities of people directly related to the use of the software are shown. Various factors that determine the interaction of the user with the software and affect his engagement are analyzed. The possibilities of using user engagement to assess competitiveness, usability, popularity and the overall quality of the software itself with which the user interacts are considered. Approaches to using the results of the user engagement assessment for adapting the learning process in learning systems are considered. The method of combined online assessment of engagement is described, in which engagement is considered as an indirect assessment of the success of the user's learning process and the results of the analysis of engagement are used to adapt the learning process to this user. The results of the study are represented as a mental map with the categorization of the concept of engagement. Directions for further research are described related to the development of an intelligent support system for assessing user engagement, based on the analysis of observed heterogeneous data on the current state of the user and behavioral metrics collected in the process of user interaction with the software and a combination of methods for their processing.