Exploring the mechanisms behind the assessment of usefulness of restaurant reviews

Abstract

Local online reviews such as Yelp have become large repositories of information, thus making it difficult for readers to find the most useful content. Our work investigates the factors that influence the readers' judgment of usefulness of restaurant reviews. We focus on assessing the mechanism behind the users' assessment of usefulness of reviews, particularly with respect to reviews provided by reviewers with local knowledge. We collected 160 manual annotations of 36 unique restaurant reviews and we interviewed ten participants. Our results show that users are able to detect reviews written by knowledgeable locals, and they perceive reviews provided by locals more useful not because they provide more valuable content but because local knowledge results in higher trust. We discuss design implications of these findings for helping readers to overcome information overload in local systems.