Computer scientists from AT&T, Princeton and other universities are at work developing a method to allow researchers to analyze location data from mobile phones in a way that doesn't use identifying data. The model works by using an algorithm to look for aggregate data sets that could identify an individual and altering them slightly. "Noise is injected into the model at points in order to reduce the likelihood of individuals being identifiable," Princeton computer scientist Margaret Martonosi said.

Related Summaries