Inferring privacy policies for social networking services

Social networking sites have come under criticism for their poor privacy protection track record. Yet, there is an inherent difficulty in deciding which principals should have access to user’s information or actions, without requiring them to constantly manage their privacy settings. We propose to extract automatically such privacy settings, based on the policy that information produced within a social context should remain in that social context, both to ensure privacy as well as maximising utility. A machine learning approach is used to extract automatically such social contexts, as well as a tentative evaluation. Categories and Subject Descriptors: K.4.1 [Computers and Society]: Public policy issues – privacy; General Terms: Security

• Type: Conference
• Year: 2009
• Where: CCS
• Authors: George Danezis
• Keywords: CCS 2009, Privacy Settings, Security Privacy, Social Contexts, Social Networking Sites