A dynamic logistic regression for network link prediction

Abstract

In social network analysis, link prediction is a problem of fundamental importance. How to conduct a comprehensive and principled link prediction, by taking various network structure information into consideration, is of great interest. To this end, we propose here a dynamic logistic regression method. Specifically, we assume that one has observed a time series of network structure. Then the proposed model dynamically predicts future links by studying the network structure in the past. To estimate the model, we find that the standard maximum likelihood estimation (MLE) is computationally forbidden. To solve the problem, we introduce a novel conditional maximum likelihood estimation (CMLE) method, which is computationally feasible for large-scale networks. We demonstrate the performance of the proposed method by extensive numerical studies.