You Are Who Knows You: Predicting Links Between Non-members of Facebook
Could online social networks like Facebook be used to infer relationships between non-members? We show that the combination of relationships between members and their e-mail contacts to non-members provides enough information to deduce a substantial proportion of the relationships between non-members. Using structural features we are able to predict relationship patterns that are stable over independent social networks of the same type. Our findings are not specific to Facebook and can be applied to other platforms involving online invitations.
KeywordsOnline social network Privacy Link prediction Machine learning Random forest classifier
E.Á. Horvát and K.A. Zweig were supported by the Heidelberg Graduate School of Mathematical and Computational Methods for the Sciences, University of Heidelberg, Germany, which is funded by the German Excellence Initiative (GSC 220). F.A. Hamprecht and K.A. Zweig were supported by a fellowship of the Marsilius Kolleg, University of Heidelberg, Germany.
- 13.Wang C, Satuluri V, Parthasarathy S (2007) Local probabilistic models for link prediction. In: 7th IEEE international conference on data mining, pp 322–331 Google Scholar