Key points are not available for this paper at this time.
Given users' growing penchant to use online reviews for travel planning, the business malpractice of posting manipulative reviews to distort the reputation of hotels is on the rise. Some manipulative reviews could be positive and intended to boost own offerings, while others could be negative and meant to slander competing ones. However, most scholarly inquiry hitherto has been trained on the former. Hence, this paper investigates the extent to which linguistic cues such as readability, genre and writing style of negative reviews could help predict if they are manipulative or authentic. Analysis of a publicly available dataset of 800 negative reviews (400 manipulative + 400 authentic) indicates that manipulative reviews are generally less readable than authentic reviews. In terms of genre, although manipulative reviews should be imaginative and authentic reviews informative, spammers appear adept enough to blur the line between the two. With respect to writing style, manipulative reviews are more richly embellished with affective cues and perceptual words.
Banerjee et al. (Thu,) studied this question.