Tell Me Something New: Selecting Novel Online Reviews

Dicle Yagmur Ozdemir, Sumit Sarkar

Research output: Working paperAcademic

Abstract

Reviews help consumers learn about the quality of a product or service. However, reading reviews can be time-consuming when there exists a large number of reviews for each product. To reduce the effort a reader expends reading enough reviews to make a decision, platforms can identify a subset of informative reviews to present to the reader. Such a subset can help the consumer obtain as much new information as possible in a short time. While platforms recognize that novel information is important for contributors, platforms do not currently provide such a facility to readers to identify the novel content easily. This paper develops an
approach that gives platforms the ability to select and present a small review set that provides as much novel information as possible. To solve this problem, a novelty formulation for a review set is presented, and some desirable properties of novelty measures are identified. The problem is difficult to solve, and efficient heuristics exploiting the structure of the problem are proposed. Experiments are conducted on restaurant reviews from Yelp to demonstrate the effectiveness of the proposed methods. The results show that the heuristics identify review sets that are very close to optimal. Heuristics designed for real-time environments are shown to be very scalable while still generating excellent solutions. The proposed approaches are very versatile—we demonstrate how they can be extended to scenarios where the selected subset of reviews preserve the average opinion and aspect coverage in a corpus while still providing high levels of novel information.
Original languageEnglish
Publication statusPublished - 2023

Fingerprint

Dive into the research topics of 'Tell Me Something New: Selecting Novel Online Reviews'. Together they form a unique fingerprint.

Cite this