Many datasets can naturally be represented as graph where nodes represent instances and links represent relationships between those instances. A fundamental problem with these types of data is that the link information in the graph may be of dubious quality; links may incorrectly exist between unrelated nodes and links may be missing between two related nodes. The goal of link prediction is to predict the existence of incorrect or missing links between the nodes of the graph.
Inferring the existences of edges between nodes in a graph has traditionally been referred to as link prediction (Liben-Nowell and Kleinberg 2003a; Taskar et al. 2003). Link prediction is a challenging problem that has been studied in various guises in different domains. For example, in social network analysis, there is work on predicting friendship links (Zheleva et al. 2008), event participation links (i.e., coauthorship O’Madadhain...
- Balasubramanyan R, Carvalho VR, Cohen W (2009) Cutonce recipient recommendation and leak detection in action. In: Workshop on enhanced messaging, ChicagoGoogle Scholar
- Diehl C, Namata GM, Getoor L (2007) Relationship identification for social network discovery. In: Proceedings of the 22nd national conference on artificial intelligence, VancouverGoogle Scholar
- Goldenberg A, Kubica J, Komarek P, Moore A, Schneider J (2003) A comparison of statistical and machine learning algorithms on the task of link completion. In: Conference on knowledge discovery and data mining, workshop on link analysis for detecting complex behavior, Washington, DCGoogle Scholar
- Liben-Nowell D, Kleinberg J (2003a) The link prediction problem for social networks. In: International conference on information and knowledge management, New OrleansGoogle Scholar
- Liben-Nowell and Kleinberg (2003b)Google Scholar
- Milne D, Witten IH (2008) Learning to link with wikipedia. In: Proceedings of the 17th ACM conference on information and knowledge management, Napa ValleyGoogle Scholar
- Popescul A, Ungar LH (2003) Statistical relational learning for link prediction. In: International joint conferences on artificial intelligence workshop on learning statistical models from relational dataGoogle Scholar
- Rattigan MJ, Jensen D (2005a) The case for anomalous link discovery. SIGKDD Explor Newsl 7:41–47Google Scholar
- Rattigan and Jensen (2005b)Google Scholar
- Szilagyi A, Grimm V, Arakaki AK, Skolnick J (2005a) Prediction of physical protein-protein interactions. Phys Biol 2(2):S1–S16Google Scholar
- Szilagyi et al. (2005b)Google Scholar
- Taskar B, Wong M-F, Abbeel P, Koller D (2003) Link prediction in relational data. In: Advances in neural information processing systems, VancouverGoogle Scholar
- Zheleva E, Getoor L, Golbeck J, Kuter U (2008) Using friendship ties and family circles for link prediction. In: 2nd ACM SIGKDD workshop on social network mining and analysis, Las VegasGoogle Scholar
- Zhu J (2003) Mining web site link structure for adaptive web site navigation and search. Ph.D. thesis, University of Ulster at JordanstownGoogle Scholar