English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

The promising performance of Deep Neural Networks (DNNs) in textclassification, has attracted researchers to use them for fraud reviewdetection. However, the lack of trusted labeled data has limited theperformance of the current solutions in detecting fraud reviews. The GenerativeAdversarial Network (GAN) as a semi-supervised method has demonstrated to beeffective for data augmentation purposes. The state-of-the-art solutionsutilize GANs to overcome the data scarcity problem. However, they fail toincorporate the behavioral clues in fraud generation. Additionally,state-of-the-art approaches overlook the possible bot-generated reviews in thedataset. Finally, they also suffer from a common limitation in scalability andstability of the GAN, slowing down the training procedure. In this work, wepropose ScoreGAN for fraud review detection that makes use of both review textand review rating scores in the generation and detection process. Scores areincorporated through Information Gain Maximization (IGM) into the loss functionfor three reasons. One is to generate score-correlated reviews based on thescores given to the generator. Second, the generated reviews are employed totrain the discriminator, so the discriminator can correctly label the possiblebot-generated reviews through joint representations learned from theconcatenation of GLobal Vector for Word representation (GLoVe) extracted fromthe text and the score. Finally, it can be used to improve the stability andscalability of the GAN. Results show that the proposed framework outperformedthe existing state-of-the-art framework, namely FakeGAN, in terms of AP by 7\%,and 5\% on the Yelp and TripAdvisor datasets, respectively.

ScoreGAN: A Fraud Review Detector based on Multi Task Learning of Regulated GAN with Data Augmentation