Detecting Fake Reviews Using Aspect-Based Sentiment Analysis and Graph Convolutional Networks DOI Creative Commons

Prathana Phukon,

Πέτρος Ποτίκας, Katerina Potika

et al.

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(7), P. 3771 - 3771

Published: March 29, 2025

Online reviews significantly influence consumer behavior and business reputations. Detecting fake is important for maintaining trust integrity in these platforms. We present an aspect-based sentiment analysis approach, referred to as FakeDetectionGCN, distinguish genuine feedback from deceptive content. The idea analyze sentiments related specific aspects (features) within reviews. Graph convolutional networks are used model the complex contextual dependencies review texts. Additionally, SenticNet, external semantic resource, integrated enhance understanding of This capable identifying both human-generated (genuine) well computer-generated (fake) It has been evaluated on two types datasets shown strong performance across both. Through this work, we contribute effective detection a trustworthy online ecosystem.

Language: Английский

Detecting Fake Reviews Using Aspect-Based Sentiment Analysis and Graph Convolutional Networks DOI Creative Commons

Prathana Phukon,

Πέτρος Ποτίκας, Katerina Potika

et al.

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(7), P. 3771 - 3771

Published: March 29, 2025

Online reviews significantly influence consumer behavior and business reputations. Detecting fake is important for maintaining trust integrity in these platforms. We present an aspect-based sentiment analysis approach, referred to as FakeDetectionGCN, distinguish genuine feedback from deceptive content. The idea analyze sentiments related specific aspects (features) within reviews. Graph convolutional networks are used model the complex contextual dependencies review texts. Additionally, SenticNet, external semantic resource, integrated enhance understanding of This capable identifying both human-generated (genuine) well computer-generated (fake) It has been evaluated on two types datasets shown strong performance across both. Through this work, we contribute effective detection a trustworthy online ecosystem.

Language: Английский

Citations

0