Social Network Analysis for privacy
25
Feb
The University of Pennsylvania, led by Dr.Sofya Raskhodnikova, associate professor of Computer Science and Engineering whose research uses the Social Network Analysis Methodology to verify the privacy on social networking media, using the information that is available in public databases.
These databases can be easily correlated data across the deleted data and recovered in user identification information operations, what is called “differential privacy.”
Differential privacy is here that restricts the type of analysis that can performed for which the presence or absence of a person is negligible, it is necessary to increase the accuracy of the analysis, avoiding the identification of individual records and thus the differential privacy is one that differentiates these two databases and provide data that should always be hidden.
The Raskhodnikova researcher states that “an approach to achieve differential privacy is the addition of a small amount of noise to the actual statistics before publishing them,” but the problem is to determine how much noise and how to run it so that the accuracy of Results are maintained.
For this design idea of the differential privacy could be especially important for the protection of data placed on a graph that identifies this data without revealing them.
Thus, the researchers say, it is possible to discover data statistics without releasing their contents, and these are differentially private methods.
The degree distribution of a social network specifies how many friends each member has, but some information is then inherently sensitive to be released with differential privacy, then protection measures can be adopted.
