Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{178086,
        author = {Deepa.S and Kanishka R and Anandhi A and Sathiyavani M and Thirisha V},
        title = {Detecting Fake Online Reviews using Random Forest with Generative Adversarial Networks},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {12},
        pages = {2813-2818},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=178086},
        abstract = {User reviews have emerged as essential in customer selection and business image building because of e-commerce development, social media, and online service growth. The widespread presence of deceptive review manipulation creates considerable problems when we aim to guarantee their trustworthiness. The detection of fraudulent reviews has become a crucial research problem because they lead customers astray while degrading business product rankings and causing financial losses. Traditional fake review detection methods, such as rule-based approaches, sentiment analysis, and conventional machine learning algorithms, face multiple disadvantages in their operation. Many present algorithms face difficulties working within extensive feature domains, which causes computational challenges and decreases their interpretability level. This study proposes integrating Regular Expression Matching (REM) for preprocessing together with Principal Component Analysis (PCA) for feature selection to use as components of a novel Random Forest with Generative Adversarial Networks (RF-GAN) classification model. REM functions to standardize textual information by eliminating extraneous characters, symbols, and normalization errors. PCA's dimensional reduction techniques maintain fundamental patterns from the data, so both computational efficiency and interpretability benefits become possible. RF-GAN merges GAN-based artificial data creation with RF ensemble performance to create a solution that handles data imbalance issues and advances classification outcomes. Results show that RF-GAN produces 95.2% accuracy as a solution for detecting fake reviews in dynamic online environments.},
        keywords = {e-commerce, user review, fake review, sentiment analysis, REM, PCA, RF-GAN},
        month = {May},
        }