Key Laboratory of Education Blockchain and Intelligent Technology, Ministry of Education Guangxi Normal University, Guilin 541004, China
2.
Guangxi Key Laboratory of Multi-source Information Mining & Security, Guangxi Normal University, Guilin 541004, China
3.
College of Computer Science and Information Engineering, Guangxi Normal University, Guilin 541004, China
Funds: The work was supported by the National Natural Science Foundation of China under Grant No. 62262003 and the Key Program of the National Natural Science Foundation of China under Grant No. U21A20474, the Guangxi Natural Science Foundation under Grant No. 2020GXNSFAA297075, the Guangxi “Bagui Scholar” Teams for Innovation and Research Project, the Guangxi Collaborative Innovation Center of Multi-source Information Integration and Intelligent Processing, the Guangxi Talent Highland Project of Big Data Intelligence and Application and the Research Fund of Guangxi Key Lab of Multi-source Information Mining & Security under Grant No. 19-A-02-02.
The Cross Domain Recommendation (CDR) technique has been widely applied in recommendation systems and can effectively relieve data sparsity and cold start problems. However, it brings up data privacy and authenticity challenges due to different data sources. Most existing solutions only con-sider data privacy, and are not robust against participant poisoning attacks that affect the quality of recommendations. How to leverage the privacy and authenticity of data remains a key challenge. To this end, we propose a blockchain-based anonymous model for CDR systems (BAM_CRS). Under this model, we use a three-chain structure to store users, products and the relationships between the two parties to guarantee privacy. This structure enables transactions to be separated from users and items to guarantee privacy and to establish consensus without relying on a central authority. For da-ta authenticity, we warrant data imtamability and traceability through on-chain data collection and design a contribution-based quantification and incentive mechanism to ensure system sustainability. Additionally, a signature-based verifiable mechanism is designed to motivate node participation and further guarantee data authenticity. Experiments are performed on real-life datasets and the results demonstrate that our model can protect privacy with an advantage of about 2% improvement over baselines in terms of precision and F1 metrics.