Helping Users Managing Context-Based Privacy Preferences

摘要

Today, users interact with a variety of online services offered by different providers. In order to supply their services, providers collect, store and process users' data according to their privacy policies. To have more control on personal data, user can specify a set of privacy preferences, encoding the conditions according to which his/her data can be used and managed by the provider. Moreover, many services are context dependent, that is, the type of delivered service is based on user contextual information (e.g., time, location, and so on). This makes more complicated the definition of privacy preferences, as, typically, users might have different attitude with respect the privacy management based on the current context (e.g., working hour, free time). To provide a more fine-grained control, a user can set up different privacy preferences for each different possible contexts. However, since user change the context very frequently, this might result in a very complex and time-consuming task. To cope with this issue, in this paper, we propose a context-based privacy management service that helps users to manage their privacy preferences setting under different contexts. At this aim, we exploit machine learning algorithms to build a classifier, able to infer new privacy preferences for the new context. The preliminary experimental results we have conducted are promising, and show the effectiveness of the proposed approach.