Understanding Consumers’ Sentiment Expressions in Online Reviews: A Hybrid Approach

Abstract

Abstract This study explores consumers’ sentiment and its influencing factors in the context of theme parks by analyzing online reviews. A hybrid approach is employed, including sentiment analysis, logistic regression analysis, and co-occurrence network analysis. The findings provide new insights on the sentiment expression of consumers in different regions, that is, Australian consumers express significantly more positive sentiment than US consumers. In addition, the results reveal the crucial experiential factors causing positive and negative sentiment. This study not only contributes to the understanding of consumers’ review behavior and eWoM but also provides a reference for the application of data driven approach in intelligent marketing and management.