Inference detection technology for Web 2.0
Consider the problem of inference control: what information can be released publicly without compromising a set of secrets. The difficulty of the problem comes from the quantity and complexity of possible inferences that might arise when released data is combined with public knowledge. This problem, and its impact on privacy, has gained increased prominence with Web 2.0 technologies that facilitate data generation and availability. We describe an approach to this problem based on using the Web itself to estimate inference likelihood. This approach is based on the reasoning that the Web encapsulates a large fraction of public knowledge and is thus an excellent proxy for that knowledge.
Chow, R. ; Golle, P. ; Staddon, J. Inference detection technology for Web 2.0. Web 2.0 Security and Privacy 2007; 2007 May 24; Oakland, CA, USA.