Detection of E-commerce Fake Reviews Using Core Diagram and Metric Weight Measurement Algorithms

In today's digital age, e-commerce platforms have become an essential part of daily life. However, the convenience and openness of online shopping has also led to numerous legal and ethical issues. Dishonest merchants, in pursuit of higher profits, often hire fake reviewers to post misleading comments, undermining consumer trust and violating trade laws. Therefore, in response to the detection of such fraudulent activities in the e-commerce environment, this study proposes a method that uses core diagrams and metric weight measurements to identify fake reviews. By evaluating the relevance of users based on rating levels and temporal correlation, a user relationship graph was constructed, which served as the basis for the detection algorithm. The method improved the accuracy of fake review detection by employing a multi-label propagation strategy and integrating an algorithm that combined entropy and analytic hierarchy process methods for metric weight measurement. The experimental setup was conducted on four real-world datasets—Amazon, YelpChi, YelpNYC, and YelpZip. The results showed that the proposed method achieved an average accuracy of 0.88, a precision of 0.88, a recall rate of 0.85, and an F1 score of 0.87 on the Amazon dataset, significantly outperforming other methods. These findings highlight the applicability and reliability of the model proposed in this study in the field of e-commerce fake review detection, providing a strong solution to protect consumer interests and maintain fair competition in the online market.