Fake review identification and utility evaluation model using machine learning

Wonil Choi; Kyungmin Nam; Minwoo Park; Seoyi Yang; Sangyoon Hwang; Hayoung Oh

doi:10.3389/frai.2022.1064371

Fake review identification and utility evaluation model using machine learning

Front Artif Intell. 2023 Jan 19:5:1064371. doi: 10.3389/frai.2022.1064371. eCollection 2022.

Authors

Wonil Choi¹, Kyungmin Nam¹, Minwoo Park¹, Seoyi Yang¹, Sangyoon Hwang¹, Hayoung Oh²

Affiliations

¹ Department of Business Administration, Sungkyunkwan University, Seoul, South Korea.
² College of Computing and Informatics, Sungkyunkwan University, Seoul, South Korea.

Abstract

Due to the structural growth of e-commerce platforms, the frequency of exchange of opinions and the number of online reviews of platform participants related to products are increasing. However, given the growth of fake reviews, the corresponding growth in the quality of online reviews seems to be slow, at best. The number of cases of harm to retailers and customers caused by malicious false reviews is steadily increasing every year. In this context, it is becoming difficult for users to determine useful reviews amid a flood of information. As a result, the intrinsic value of online reviews that reduce uncertainty in pre-purchase decisions is blurred, and e-commerce platforms are on the verge of losing credibility and traffic. Through this study, we intend to present solutions related to review filtering and classification by constructing a model for judging the authenticity and usefulness of online reviews using machine learning.

Keywords: SVC; e-commerce; fake review; fake review detection technique; logistic regression; machine learning; useful reviews.

Grants and funding

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea Government (MSIT) (No. NRF-2022R1F1A1074696).