A novel approach to predict the helpfulness of online reviews

Abstract

Online reviews help consumers reduce uncertainty and risks faced in purchase decision making by providing information about products and services. However, the overwhelming amount of data continually being produced in online review platforms introduce a challenge for customers to read and judge the reviews. This research addresses the problem of misleading and overloaded information by developing a novel approach to predict the helpfulness of online reviews. The proposed approach in this study, first, clusters reviews using reviewer-related, and temporal factors. It then uses review-related factors to predict online review helpfulness in each cluster. Using a sample of Amazon.com reviews, the empirical findings offer strong support to the proposed approach and show its superior predictions of review helpfulness compared to earlier approaches. The outcomes of this study help customers in online shopping and assist online retailers in reducing information overload to improve their customers' experience.