A difference of multimedia consumer’s rating and review through sentiment analysis

With the increase in communication using online and mobile channels, consumers, in purchasing products or services, use online user reviews as an important decision-making tool providing information about other consumers’ experiences. Companies also objectively analyze consumer opinions about their products and services to derive business insights. However, the rating, a quantitative information that serves as the basis for the evaluation and recommendation system of reviews, has the limitation that it does not reflect actual consumers’ opinions or is inappropriate for use in recommendation systems. Sentiment analysis, which identifies the emotions and feelings contained in the online text generated by users, has been proposed as a way to deal with and solve such problems related to existing ratings. The present study aimed to investigate, using the lexicon-based approaches for review sentiment analysis, to whether the rating systems reflect genuine experience, satisfaction, and thoughts of customers. The results of this study demonstrate that reviews do not accurately represent the positive or negative aspects of the reviewed products and services, suggesting the need for future modifications in terms of system and consumer analysis. This study also suggests directions of future system and consumer research.