Emotions in Online Reviews and Product Sales: Unifying Empirical and Theoretical Perspectives

Abstract

Emotion artificial intelligence, the algorithm that recognizes and interprets various human emotions beyond valence (positive and negative polarity), is still in its infancy yet attracts much attention from both the industry and the academia. Based on the discrete emotion theory and statistical language modeling, this work proposes an algorithm to enable automatic domain-adaptive emotion lexicon construction and multi-dimensional emotion detection in texts. With a large-scale dataset of China’s movie market from 2012 to 2018, we construct and validate a domain-specific emotion lexicon and demonstrate the predictive power of eight discrete emotions (i.e., surprise, joy, anticipation, love, anxiety, sadness, anger, and disgust) in online reviews on box office. Further, we find that representing emotions using discrete emotions yields higher prediction accuracy than using valence or latent emotion variables generated by topic modeling. To understand the source of the predictive power from a theoretical perspective, and to test the cross-culture generalizability of our prediction study, we further conduct an experiment in the U.S. movie market, based on the “feelings as information” theory and theoretical research on emotion, judgement, and decision making. We find that mediated by processing fluency, discrete emotions significantly affect the perceived review helpfulness, which further influences purchase intention. Our work shows the economic value of emotions in online reviews, generates insight into the mechanism of their effects, and has managerial implications for online review platform design, movie marketing, and cinema operations.