Modeling Psychological Triggers of FoMO in Digital Media Using Machine Learning Approaches

Abstract

FoMO (Fear of Missing Out), as a generic psychological trigger is proposed as an account for the mechanism of user attention, decision and behavior in digital pervasive environments where more and more cues of urgency, social proof and scarcity appear as integral parts in contents. In this paper, a multimodal machine learning model for operationalizing the representative psychological triggers of FoMO and behavior prediction based on digital pervasive ad engagement data is put forward. The hybrid classifier ensemble combining semantics from language-related urgency and semantics from visual-content scarcity and temporal engagement predicts their corresponding emotion trigger of FoMO instantiation via explainable AI techniques. The proposed multimodal approach gets the best performance of 92.67% in accuracy and 0.962 in AUC among the comparison with existing traditional machine learning models, which enhances the prediction accuracy using visual and language cues about FoMO.