Mitigating Review and Rating Fraud in E-Commerce Platforms: A Blockchain-Based Reputation System with AI-Driven Review Validation
نویسندگان
1 Faculty of Computer Engineering, University of Isfahan, Iran.
2 Faculty of Computer Engineering, University of Isfahan, Iran.
3 Faculty of Computer Engineering, University of Isfahan, Iran.
doi
10.22108/jcs.2025.143363.1154چکیده
In the world of e-commerce, the impact of reviews on buyers’ decision-making is undeniable. These reviews serve as a measure of product quality and an indicator of sellers' credibility. Current platforms receive and store a vast number of reviews from customers. However, these systems are often vulnerable to rating fraud, as dishonest buyers and sellers exploit them by intentionally promoting or discrediting specific products. Additionally, the lack of sufficient oversight on these platforms can lead to the misuse of reviews by the platform itself to boost sales. Existing methods for combating these issues frequently rely on centralized systems that lack transparency and are susceptible to manipulation. This creates a negative experience for online buyers and significantly reduces trust in e-commerce platforms. This research presents a new method for addressing the issue of review and rating fraud and supporting effective reward allocation to customers using blockchain and smart contract technologies. By creating a decentralized and immutable system, blockchain minimizes fraud and manipulation in reviews, restoring buyers' trust in e-commerce systems. Additionally, this paper presents a decentralized artificial intelligence model that identifies genuine and fake reviews, incentivizing active and honest users through a reward mechanism. Results indicate that the proposed mechanism addresses more challenges compared to other methods while maintaining desirable features of decentralization, data security, effective reward allocation, and reputation protection.