Do topic consistency and linguistic style similarity affect online review helpfulness? An elaboration likelihood model perspective

A critical assessment of consumer reviews: A hybrid NLP-based methodology

“…However, the uniqueness of information as perceived by a user when reading multiple reviews can also vary with their semantic themes [68]. For instance, when comprehending the information coverage of two similar reviews, "the quality of printing is very nice and gives more quality printing than other refill inks" and "highly recommended printer cartridge product," the user analyses their information richness in combination with their semantic content and linguistic style of presentation [25].…”

“…Petty and Cacioppo [50] proposed the Elaboration Likelihood Model as a form of the Dual Process Theory. ELM Theory suggests two primary routes to persuasion with the help of communication cues while explaining the helpfulness of OCRs [1,10,45,68,69]. The first route involves thoughtful consideration of central cues directly related to an OCR's helpfulness.…”

“…Online reviews, also known as electronic word-of-mouth (e-WOM), are user-generated contents that enable customers to present their first-hand assessments about products and services after consumption. Most users treat online reviews as personal recommendations based on first-hand usage experiences from previous buyers and acknowledge them to be more effective than traditional marketing communication mix while describing the pros and cons of a product or service [1,10,18,45,68,69]. According to the 2021 Statista Global Consumer Survey conducted in the United States, 40 per cent of buyers consulted online reviews to search product-specific information, ranking second only after search engines (measured at 62 per cent) 1 .…”

Towards explaining user satisfaction with contact tracing mobile applications in a time of pandemic: a text analytics approach

“…Second, we used text analysis and the WordCloud library in Python to visualize the most common words in reviews (see Figure 4). The most frequent words (and their frequency) included: phone (3070), number (2445), work (1982) Third, we applied topic modelling [50] using LDA (Latent Dirichlet Allocation) and extracted topics from the whole review texts. We then followed an abductive reasoning approach.…”

Predicting review helpfulness in the omnichannel retailing context: An elaboration likelihood model perspective

“…Online customer reviews play an increasingly important role in online commerce which has attracted scholars and practitioners’ attention around the world ( Yang et al, 2021 ; Li et al, 2022 ; Wu et al, 2022 ). Extensive numbers of product reviews provide rich information for consumers which can help them decrease the degree of uncertainty and risk ( Zhu and Zhang, 2010 ; Yang et al, 2019 ).…”