Link Prediction in Multiplex Network Based on Regression and Conditional Probability

Multiplex networks are often used to describe the relationship of different properties between the same group of entities in real complex system, in which nodes represent entities and links in different layers represent connections of different properties between entities. The key to link prediction in multiplex networks lies in (1) making full use of the information provided by each layer of a network; (2) effectively fusing the information provided by each layer of the network together. In this paper, we propose a method based on regression and conditional probability, called MRCP, in which the feature vectors of node pairs are the vectors proposed in our previous work. This method combines intralayer probability and interlayer information to predict missing links in multiplex networks. Firstly, the intralayer probability is calculated by using regression algorithm based on intralayer information. Then the conditional probability of link existence is calculated by using the auxiliary layer information. Finally, both probabilities are combined for link prediction. In order to verify the effectiveness of the method, we conducted experiments on 8 real datasets. The experimental results show that the prediction performance of this method is better than compared methods.