A Study on the User Privacy-Preserving Personalized Recommendation Based on Random Perturbation for online Services in the Background of Big Data

In the background of big data, how to realize personalized online service recommendation while protecting user privacy is the current research hotspot. This paper first analyzes the collaborative filtering algorithm and the user privacy-preserving problems of the online service personalized recommendation on the big data background, and then introduces the user personal characteristic model in the collaborative filtering recommendation to improve the accuracy of the recommendation and the user privacy is preserved at the same time. Considering that the current privacy-preserving method does not reflect user participation and balance between privacy preserving and recommendation accuracy, this paper improves the privacy-preserving method through user preference policy, and uses distributed cloud storage for data storage and processing to further prevent malicious attacks by attackers. Finally, the effectiveness of the improved random perturbation-based method is demonstrated through data comparative analysis. Experiments demonstrate that the method proposed in this paper effectively achieves privacy protection while ensuring a high recommendation accuracy of personalized services.