Online reviews help consumers make purchasing decisions on e-commerce platforms by addressing uncertainties in online shopping. In the context of information overload, the value of review helpfulness has become increasingly evident. This study aims to explore the effects of discrete emotions on helpfulness by categorizing Ekman’s six basic emotions based on arousal and valence, and extracting their probabilities from Amazon reviews using BERT (Bidirectional Encoder Representations from Transformers). The results indicate high arousal or negative emotions adversely influence review usefulness. Furthermore, anger, disgust, and fear, characterized by high arousal and displeasure, significantly affect the perceived helpfulness of the reviews. Additionally, we identify product involvement as a significant moderator, suggesting that consumers’ investment in products influences their interpretation of reviews’ emotional content. This study contributes to understanding the relationship between discrete emotions, product involvement, and review helpfulness in e-commerce, offering insights for online retailers and platform developers to enhance consumer decision-making.