User Sensitive Subject Protection Method For Book Search Service

Book search service is one of the most important information services provided by digital libraries to users.However,with the rapid development of emerging technologies such as big data,the server side of digital libraries is becoming more and more untrustworthy.How to prevent the leakage of user book query privacy has attracted widespread attention.This paper proposes a user-based client,based on the user's historical book query sequence,to construct a set of pseudo book query sequence sets highly similar to the user's real book query sequence feature distribution.The theme of the pseudo book covers the sensitive theme contained in the user's book query.First of all,this article presents a book search privacy protection framework composed of an untrusted server and a set of trusted clients.This framework does not need to change the server-side book search algorithm or affect the accuracy of the book search service.Then,this paper takes the book query sequence as the research unit,defines the user query privacy protection model,and formalizes the constraints that the ideal pseudo query should meet:(1)It has a highly similar feature distribution to the user's real query to achieve the goal of "real and fake are difficult to identify";(2)can effectively reduce the saliency of usersensitive topics on the untrusted server side,so as to achieve the purpose of confusing user-sensitive topics.Third,based on the above privacy protection model,this paper implements a specific algorithm for the privacy protection model of sensitive topics,and generates a set of pseudo book query sequences that satisfy the model constraints for the user's real book query sequence.Fourth,this paper verifies the effectiveness of this method through theoretical analysis and experimental evaluation.Finally,according to the experimental results,this paper compares the effectiveness of the method with other existing privacy protection technologies from the four dimensions of privacy security,service accuracy,service efficiency,and actual availability.Compared with other methods,this method has more Excellent overall performance.In summary,the work in this article is a brand-new attempt for privacy-related issues of user-sensitive topics in book search services,which achieves the actual availability of the book search service of the digital library,the accuracy of user book query results and the digital library On the premise of the high efficiency of the book search service performance,it ensures the security of the user's book-sensitive topics on the server side of the digital library.It is hoped that it will have a positive and effective significance for the research of a more comprehensive and effective digital library user privacy protection algorithm.