Cold-start link prediction in multi-relational networks based on network dependence analysis

Abstract Cold-start link prediction has been a hot issue in complex network. Different with most of existing methods, this paper utilizes multiple interactions to predict a specific type of links. In this paper, multiple interactions are abstracted as multi-relational networks, and robust principle component analysis is employed to extract low-dimensional latent factors from sub-networks. Then a distribution free independence test, projection correlation, is introduced to efficiently analyze dependence between target and auxiliary sub-networks. Furthermore, associated auxiliary networks are exploited for cold-start link prediction, which aims to forecast potential links for new/isolated nodes in target sub-networks. Experimental results on 8 bioinformatics datasets validate rationality and effectiveness of the method.

[1]  Linyuan Lu,et al.  Link Prediction in Complex Networks: A Survey , 2010, ArXiv.

[2]  Jianhua Ruan,et al.  A novel link prediction algorithm for reconstructing protein-protein interaction networks by topological similarity , 2013, Bioinform..

[3]  Bin Li,et al.  Link prediction in multi-relational networks based on relational similarity , 2017, Inf. Sci..

[4]  Hong Cheng,et al.  Link prediction via matrix completion , 2016, ArXiv.

[5]  M S Waterman,et al.  Identification of common molecular subsequences. , 1981, Journal of molecular biology.

[6]  M. Kanehisa,et al.  Development of a chemical structure comparison method for integrated analysis of chemical and genomic information in the metabolic pathways. , 2003, Journal of the American Chemical Society.

[7]  Runze Li,et al.  Projection correlation between two random vectors , 2017, Biometrika.

[8]  Jung Yeol Kim,et al.  Correlated multiplexity and connectivity of multiplex random networks , 2011, 1111.0107.

[9]  Giorgio Fagiolo,et al.  Multinetwork of international trade: a commodity-specific analysis. , 2009, Physical review. E, Statistical, nonlinear, and soft matter physics.

[10]  Qi Zhang,et al.  Cold-start link prediction in multi-relational networks , 2017 .

[11]  Linyuan Lu,et al.  Uncovering missing links with cold ends , 2011, ArXiv.

[12]  Vito Latora,et al.  Measuring and modelling correlations in multiplex networks , 2014, Physical review. E, Statistical, nonlinear, and soft matter physics.

[13]  Linyuan Lü,et al.  Predicting missing links via local information , 2009, 0901.0553.

[14]  Eduardo R. Hruschka,et al.  Simultaneous co-clustering and learning to address the cold start problem in recommender systems , 2015, Knowl. Based Syst..