摘要： 在数字图书馆信息发布中用户身份信息是进行数据分析的重要依据。建立一种保留用户身份的匿名数据发布模型，对用户属性进行合理化选择，采用匿名方式划分用户身份得到用户类，通过用户类与借阅信息的链接得到用户身份与借阅信息的映射关系，从而发布包含用户身份的匿名数据。实验结果表明，保留用户身份的匿名发布能在保障用户安全的前提下，较好地实现数字图书馆数据的可利用性，整体提高数字图书馆的服务质量。

关键词: 数字图书馆, 数据发布, 隐私保护, k-匿名, 微聚集

Abstract: In the information publishing of digital library,user identity information is the important basis for data analysis.This paper builds an anonymous data publishing,makes reasonable selection for user attributes,and partitions the user identity to get the user class by anonymous way.Through connection of user classes and borrowing information,the mapping relationship between user identity and borrowing information is obtained,thus releasing the anonymous data which retains user identity.Experimental results show that under the premise of user security,the anonymity publishing which retains user identity can better realize usefulness of digital library data,and improve the quality of service of digital library as a whole.

Key words: digital library, data publishing, privacy protection, k-anonymity, micro-aggregation

中图分类号: 

• TP391