The polarising effect of Review Bomb

04/02/2021
by   Venera Tomaselli, et al.
0

This study discusses the Review Bomb, a phenomenon consisting of a massive attack by groups of Internet users on a website that displays users' review on products. It gained attention, especially on websites that aggregate numerical ratings. Although this phenomenon can be considered an example of online misinformation, it differs from conventional spam review, which happens within larger time spans. In particular, the Bomb occurs suddenly and for a short time, because in this way it leverages the notorious problem of cold-start: if reviews are submitted by a lot of fresh new accounts, it makes hard to justify preventative measures. The present research work is focused on the case of The Last of Us Part II, a video game published by Sony, that was the target of the widest phenomenon of Review Bomb, occurred in June 2020. By performing an observational analysis of a linguistic corpus of English reviews and the features of its users, this study confirms that the Bomb was an ideological attack aimed at breaking down the rating system of the platform Metacritic. Evidence supports that the bombing had the unintended consequence to induce a reaction from users, ending into a consistent polarisation of ratings towards extreme values. The results not only display the theory of polarity in online reviews, but them also provide insights for the research on the problem of cold-start detection of spam review. In particular, it illustrates the relevance of detecting users discussing contextual elements instead of the product and users with anomalous features.

READ FULL TEXT

page 2

page 11

page 13

research
05/22/2018

Estimating the Rating of Reviewers Based on the Text

User-generated texts such as reviews and social media are valuable sourc...
research
04/21/2020

Quarantine Deceiving Yelp's Users by Detecting Unreliable Rating Reviews

Online reviews have become a valuable and significant resource, for not ...
research
06/10/2020

DFraud3- Multi-Component Fraud Detection freeof Cold-start

Fraud review detection is a hot research topic inrecent years. The Cold-...
research
07/10/2022

Parametric Empirical Bayes for Predicting Quality in Rating Systems

User-solicited ratings systems in online marketplaces suffer from a cold...
research
04/13/2020

Detecting and Characterizing Extremist Reviewer Groups in Online Product Reviews

Online marketplaces often witness opinion spam in the form of reviews. P...
research
11/19/2015

BIRDNEST: Bayesian Inference for Ratings-Fraud Detection

Review fraud is a pervasive problem in online commerce, in which fraudul...
research
06/14/2018

Cold-Start Aware User and Product Attention for Sentiment Classification

The use of user/product information in sentiment analysis is important, ...

Please sign up or login with your details

Forgot password? Click here to reset