No abstract

No abstract

A new opinion extraction method is proposed to summarize unstructured, user-generated content (i.e., online customer reviews) in the fixed topic domains. To differentiate the current approach from other opinion extraction approaches, which are often exposed to a sparsity problem and lack of sentiment scores, a confirmatory aspect-based opinion mining framework is introduced along with its practical algorithm called DiSSBUS. In this procedure, 1) each customer review is disintegrated into a set of clauses; 2) each clause is summarized to bi-terms-a topic word and an evaluation word-using a part-ofspeech (POS) tagger; and 3) each bi-term is matched to a pre-specified topic relevant to a specific domain. The proposed processes have two primary advantages over existing methods: 1) they can decompose a single review into a set of bi-terms related to pre-specified topics in the domain of interest and, therefore, 2) allow identification of the reviewer's opinions on the topics via evaluation words within the set of bi-terms. The proposed aspect-based opinion mining is applied to customer reviews of restaurants in Hawaii obtained from TripAdvisor, and the empirical findings validate the effectiveness of the method.

User reviews are now an essential source of information for consumers, exerting strong influence on purchase decisions. Broadly speaking, reviews rated by consumers as more helpful exert a greater influence downstream. The current research examines how the linguistic characteristics of a review affect its helpfulness score. Using a convolutional neural network (CNN), this research analyzes the linguistic subjectivity and objectivity of over 2 million reviews on Amazon. The results show that, ceteris paribus, both linguistic subjectivity and objectivity have a positive impact on review helpfulness. However, contrary to consumers' intuition, when subjectivity and objectivity are combined in the same review, review helpfulness increases less than their respective separate effects would predict, especially for hedonic products. We conceptualize that this results from the increased complexity of messages mixing subjective and objective sentences, which requires more effortful processing. The findings extend the literature on online reviews, word‐of‐mouth, and text analysis in marketing, and offer practical implications for marketing communication and facilitation of reviews.