부산대학교 도서관

Since customers increasingly rely on online reviews from online platforms for information about products or services. The reliability of reviews becomes crucial. However, manipulated reviews give an untruthful picture of product quality, impede reviews’ usefulness, and have an impact on customers’ decision-making. Therefore, the detection of manipulated reviews has received widespread research attention. One of the main problems in detecting manipulated reviews is the difficulties with obtaining manipulated reviews and an insufficient number of manipulated reviews. In this research, we address the generation and detection of manipulated reviews by using AI. The aim of our research is to improve the detection of manipulated reviews by filling the lack of manipulated reviews with generated reviews.