A Novel Ensemble Model for Link Prediction in Social Network

Link prediction is to discover the missing connection between any pair of individuals, which could help researcher get the whole picture of social networks and observe the unknown trend or information. Based on the similarity calculation, prior studies could be categorized into two types: local information orientation and global measurement orientation. The former can use less information to obtain prediction results, while the latter requires a large amount of calculation to obtain the global information, but it has a more accurate prediction. In this study, we propose a novel ensemble model which could combine the efficiency of local information and the accuracy of global measurement to effectively predict the linkage in a huge network. The performance evaluations are conducted on multiple real datasets to demonstrate the effectiveness and practicability of the developed models. Experimental results show that our method can only use the local information of the network to achieve performance similar to the global model.