Abstract: Personalized k-anonymous models provide privacy protection based on user privacy preferences, giving users more control of location privacy, but due to the complexity of setting privacy preferences, even the most privacy-conscious users may ignore some of the problems. The purpose of this paper is to develop a framework to help users choose their own privacy preferences, effective management and access from the anonymous privacy content. Analyze a set of options that affect privacy configuration and build an adaptive learning model to help users make the right decisions and protect their privacy information. As the learning model matures, it will manage the user’s different privacy preferences in a variety of situations with minimal user intervention, prevent privacy disclosure, and encourage users to use the model’s recommended privacy settings.

Key words: adaptive learning model, location service（LBS）, privacy protection, k-anonymous

摘要： 个性化k-匿名模型能够根据用户隐私偏好实现隐私保护，为用户提供控制位置隐私更多选择性，但由于设置隐私偏好的复杂性，就算最为注重隐私保护的用户也可能忽略一些问题。研究的目的是开发一个框架，帮助用户选择自己的隐私偏好，有效管理和获取来自匿名者的隐私内容。分析一组影响隐私配置选择因素，构建自适应学习模型来帮助用户做出正确的决定，保护他们的隐私信息。随着学习模型的成熟，将以最小的用户干预来管理各种情况下不同用户的隐私偏好，防止隐私泄露，并鼓励用户使用模型推荐的隐私设置。

关键词: 自适应性学习模型, 位置服务（LBS）, 隐私保护, k-匿名