Emotion assessment and application in human–computer interaction interface based on backpropagation neural network and artificial bee colony algorithm

Abstract

Emotion assessment is a challenging task in the human–computer interaction interface. Previous studies have examined the relationship between emotion and color, but they fail to accurately analyze emotional semantics due to the numerous elements in human–computer interaction interfaces. As a result, a combination model of a backpropagation neural network (BPNN) and an artificial bee colony algorithm (ABC) was presented in this paper to predict the emotion semantics of the human–computer interaction interface. The mechanism of generating the weights and thresholds for each layer of BPNN was converted to the search for an optimal honey source. Meanwhile, according to experiment results and evaluation of elements in human–computer interaction interfaces, this paper has assessed the relationships amongst the eight key elements (ratio of graphics to text, color difference, color distribution, color harmony, theme style, white space ratio, frame style, number of colors) and emotion word pairs (moderation-fancy, calm-pleasure, confusing-clear, cold-kind, coarse-elegant). Furthermore, an emotion application database was established to determine how the amalgamation of critical elements affects the users’ feelings about the human–computer interaction interface to help designers build a user-centric interface. Finally, the database can be applied to relieve mental health problems by meeting the psychological expectations of users as mental healthcare intervention during the COVID-19 pandemic. Also, it can help designers to design a pleasurable visual interaction interface for a particular element to convey health-related information and protective measures.