A Systematic Literature Review of Computational Methods for Fake Review Detection in E-Commerce Platforms
Algorithms · 15.11.2026 · 19 (11) · 1005
DOI: 10.3390/a19111005 ↗Kurz gefasst
Systematische Übersicht in der Zeitschrift Algorithms, online erschienen am 15.11.2026. Verfasst von N. Umamaheswari und 4 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Increased importance of users’ reviews for decision-making has also occurred as a result of the emergence of e-commerce websites and because it has become acceptable to obtain information about products and services through user reviews. As a result, there has been a boom in fake reviews that manipulate the buyer’s perception of purchased goods in order to promote specific items on the Internet. Fake reviews may damage the reputation of a small company that is just entering the market, while large corporations with a high rating of their goods can simply eliminate any competition. Given the increasing need to eliminate fake reviews, machine learning (ML)-based solutions, and especially hybrid approaches and deep learning (DL) algorithms are currently being explored. This review paper systematically examines the fake review detection research from 2018 to 2025. The initial research has applied mostly supervised learning methods with high-level features extracted from the text and sentiment analysis using traditional TF-IDF techniques. However, deep learning methods such as CNN, LSTM, and BERT have gained ground in the last few years and shown better efficiency due to the ability to exploit the implicit semantics of the text. The key problems with detecting fake reviews are: unbalanced labels, a lack of multilingual and multi-culture datasets, and domain-specific performance variation for detecting fake reviews in healthcare, finance, hospitality, and consumer electronics. This survey provides an overview of the research progress in detecting fake reviews with a particular emphasis on the transition from traditional ML-based approaches to DL-based approaches and highlights key challenges and opportunities for future research in this area.
Zitieren
APA
Umamaheswari, N., Andrushia, D., Neebha, M., Nammalvar, A., & Dewi, C. (2026). A Systematic Literature Review of Computational Methods for Fake Review Detection in E-Commerce Platforms. Algorithms, 19(11), 1005. https://doi.org/10.3390/a19111005
Vancouver
Umamaheswari N, Andrushia D, Neebha M, Nammalvar A, Dewi C. A Systematic Literature Review of Computational Methods for Fake Review Detection in E-Commerce Platforms. Algorithms. 2026;19(11):1005. doi:10.3390/a19111005
BibTeX
@article{Umamaheswari2026,
author = {Umamaheswari, N. and Andrushia, Diana and Neebha, Mary and Nammalvar, Anand and Dewi, Christine},
title = {A Systematic Literature Review of Computational Methods for Fake Review Detection in E-Commerce Platforms},
journal = {Algorithms},
year = {2026},
volume = {19},
number = {11},
pages = {1005},
doi = {10.3390/a19111005}
}
EUVIMED