Correlating personal information between DOD411, LINKEDIN, FACEBOOK, and MYSPACE with uncommon names

Abstract

It is generally easier to disambiguate people with uncommon names than people with common names; in the extreme case a name can be so uncommon that it is used by only a single person on the planet, and no disambiguation is necessary. This thesis explores the use of uncommon names to correlate identity records stored in DoD411 with user profile pages stored on three popular social network sites: LinkedIn, Facebook, and MySpace. After grounding the approach in theory, a working correlation system is presented. We then statistically sample the results of the correlation to infer statistics about the use of social network sites by DoD personnel. Among the results that we present are the percentage of DoD personnel that have Facebook pages; the ready availability of information about DoD families from information that DoD personnel have voluntarily released on social network sites; and the availability of information related to specific military operations and unit deployments provided by DoD members and their associates on social network sites. We conclude with a brief analysis of the privacy and policy implications of this work.