Analysis of Facebook Friends Using Disclosure Level

Social Networking Services (SNSes), such as Facebook and Twitter, provide a communication platform that easily communicates with a family, close friends, acquaintances and strangers. On the other hands, SNS users often unintentionally post a message on SNSes, including one´s own sensitive information, and is revealed to other users. As a result, an SNS user can lose friend, or even a job. Moreover offhand posts can result in unforeseen problems. In order to prevent privacy leaks to SNSes, we defined an objective classification-table of invasion of privacy for SNSes that will be based on classification according to contents of information, which should be kept private, and disclosure level. We also proposed a method that uses the classification table to provide settings for access control that prevent privacy leaks. In this paper, as a realization of this method, we developed a prototype system that adaptive disclosure control system on Facebook based on detection of sensitive information leaks and friends classified according to the disclosure level.